posa2

Pattern-Oriented Software ArchitecturesPatterns & Frameworks for Concurrent & Distributed Systems : 

Pattern-Oriented Software Architectures Patterns & Frameworks for Concurrent & Distributed Systems Dr. Douglas C. Schmidt d.schmidt@vanderbilt.edu www.dre.vanderbilt.edu/~schmidt/ Professor of EECS Vanderbilt University Nashville, Tennessee

Tutorial Motivation: 

Tutorial Motivation Stand-alone Architecture Building robust, efficient, & extensible concurrent & networked applications is hard e.g., we must address many complex topics that are less problematic for non-concurrent, stand-alone applications Observations Fortunately, there are reusable solutions to many common challenges, e.g.: Connection mgmt & event demuxing Service initialization Error handling & fault tolerance Flow & congestion control Distribution Concurrency, scheduling, & synchronization Persistence Key challenge: How can we reuse the these solutions effectively in a changing, heterogeneous world?

Tutorial Outline: 

Tutorial Outline OO techniques & language features: Patterns (25+), which embody reusable software architectures & designs Frameworks & components, which embody reusable software middleware & application implementations OO language features, e.g., classes, dynamic binding & inheritance, parameterized types Cover OO techniques & language features that enhance software quality

Technology Trends (1/4): 

Technology Trends (1/4) Information technology is being commoditized i.e., hardware & software are getting cheaper, faster, & (generally) better at a fairly predictable rate These advances stem largely from standard hardware & software APIs & protocols, e.g.:

Technology Trends (2/4): 

Technology Trends (2/4) Growing acceptance of a network-centric component paradigm i.e., distributed applications with a range of QoS needs are constructed by integrating components & frameworks via various communication mechanisms

Technology Trends (3/4): 

Technology Trends (3/4) Components encapsulate application “business” logic Components interact via ports Provided interfaces, e.g.,facets Required connection points, e.g., receptacles Event sinks & sources Attributes Containers provide execution environment for components with common operating requirements Components/containers can also Communicate via a middleware bus & Reuse common middleware services Component middleware is maturing & becoming pervasive …

Slide7: 

e.g., standard technologies are emerging that can: Model Analyze Synthesize & optimize Provision & deploy multiple layers of QoS-enabled middleware & applications These technologies are guided by patterns & implemented by component frameworks Partial specialization is essential for inter-/intra-layer optimization <CONFIGURATION_PASS> <HOME> <…> <COMPONENT> <ID> <…></ID> <EVENT_SUPPLIER> <…events this component supplies…> </EVENT_SUPPLIER> </COMPONENT> </HOME> </CONFIGURATION_PASS> Goal is not to replace programmers per se – it is to provide higher-level domain-specific languages for middleware developers & users Model driven development is integrating generative software technologies with QoS-enabled component middleware Technology Trends (4/4)

The Evolution of Middleware: 

The Evolution of Middleware There are multiple COTS middleware layers & research/business opportunities Historically, mission-critical apps were built directly atop hardware Tedious, error-prone, & costly over lifecycles Standards-based COTS middleware helps: Control end-to-end resources & QoS Leverage hardware & software technology advances Evolve to new environments & requirements Provide a wide array of reusable, off-the-shelf developer-oriented services There are layers of middleware, just like there are layers of networking protocols Hardware Applications

Operating System & Protocols: 

Operating System & Protocols Operating systems & protocols provide mechanisms to manage endsystem resources, e.g., CPU scheduling & dispatching Virtual memory management Secondary storage, persistence, & file systems Local & remote interprocess communication (IPC) OS examples UNIX/Linux, Windows, VxWorks, QNX, etc. Protocol examples TCP, UDP, IP, SCTP, RTP, etc.

Host Infrastructure Middleware: 

Host Infrastructure Middleware Host infrastructure middleware encapsulates & enhances native OS mechanisms to create reusable network programming components These components abstract away many tedious & error-prone aspects of low-level OS APIs Domain-Specific Services Common Middleware Services Distribution Middleware Host Infrastructure Middleware

Distribution Middleware: 

Distribution Middleware Distribution middleware defines higher-level distributed programming models whose reusable APIs & components automate & extend native OS capabilities Domain-Specific Services Common Middleware Services Distribution Middleware Host Infrastructure Middleware Distribution middleware avoids hard-coding client & server application dependencies on object location, language, OS, protocols, & hardware

Common Middleware Services: 

Common Middleware Services Common middleware services augment distribution middleware by defining higher-level domain-independent services that focus on programming “business logic” Domain-Specific Services Common Middleware Services Distribution Middleware Host Infrastructure Middleware

Domain-Specific Middleware: 

Domain-Specific Middleware Domain-specific middleware services are tailored to the requirements of particular domains, such as telecom, e-commerce, health care, process automation, or aerospace Domain-Specific Services Common Middleware Services Distribution Middleware Host Infrastructure Middleware

Consequences of COTS & IT Commoditization: 

More emphasis on integration rather than programming Increased technology convergence & standardization Mass market economies of scale for technology & personnel More disruptive technologies & global competition Lower priced--but often lower quality--hardware & software components The decline of internally funded R&D Potential for complexity cap in next-generation complex systems Consequences of COTS & IT Commoditization Not all trends bode well for long-term competitiveness of traditional R&D leaders Ultimately, competitiveness depends on success of long-term R&D on complex distributed real-time & embedded (DRE) systems

Why We are Succeeding Now: 

Why middleware-centric reuse works Hardware advances e.g., faster CPUs & networks Software/system architecture advances e.g., inter-layer optimizations & meta-programming mechanisms Economic necessity e.g., global competition for customers & engineers Why We are Succeeding Now Recent synergistic advances in fundamental technologies & processes:

Example:Applying COTS in Real-time Avionics: 

Example: Applying COTS in Real-time Avionics Key System Characteristics Deterministic & statistical deadlines ~20 Hz Low latency & jitter ~250 usecs Periodic & aperiodic processing Complex dependencies Continuous platform upgrades Goals Apply COTS & open systems to mission-critical real-time avionics

Example:Applying COTS to Time-Critical Targets: 

Example: Applying COTS to Time-Critical Targets

Example: Applying COTS to Large-scale Routers: 

Example: Applying COTS to Large-scale Routers Goal Switch ATM cells + IP packets at terabit rates Key System Characteristics Very high-speed WDM links 102/103 line cards Stringent requirements for availability Multi-layer load balancing, e.g.: Layer 3+4 Layer 5 www.arl.wustl.edu

Example: Applying COTS to Software Defined Radios: 

Example: Applying COTS to Software Defined Radios Key Software Solution Characteristics Transitioned to BAE systems for the Joint Tactical Radio Systems Programmable radio with waveform-specific components Uses CORBA component middleware based on ACE+TAO www.omg.org/docs/swradio

Example:Applying COTS to Hot Rolling Mills: 

Example: Applying COTS to Hot Rolling Mills Goals Control the processing of molten steel moving through a hot rolling mill in real-time System Characteristics Hard real-time process automation requirements i.e., 250 ms real-time cycles System acquires values representing plant’s current state, tracks material flow, calculates new settings for the rolls & devices, & submits new settings back to plant

Example:Applying COTS to Real-time Image Processing: 

Example: Applying COTS to Real-time Image Processing Goals Examine glass bottles for defects in real-time System Characteristics Process 20 bottles per sec i.e., ~50 msec per bottle Networked configuration ~10 cameras www.krones.com

Key Opportunities & Challenges in Concurrent Applications: 

Motivations Leverage hardware/software advances Simplify program structure Increase performance Improve response-time Key Opportunities & Challenges in Concurrent Applications Accidental Complexities Low-level APIs Poor debugging tools Inherent Complexities Scheduling Synchronization Deadlocks

Key Opportunities & Challenges in Networked & Distributed Applications: 

Key Opportunities & Challenges in Networked & Distributed Applications Motivations Collaboration Performance Reliability & availability Scalability & portability Extensibility Cost effectiveness Accidental Complexities Algorithmic decomposition Continuous re-invention & re-discovery of core concepts & components Inherent Complexities Latency Reliability Load balancing Causal ordering Security & information assurance

Overview of Patterns: 

Overview of Patterns

Overview of Pattern Languages: 

Overview of Pattern Languages Benefits of Pattern Languages Define a vocabulary for talking about software development problems Provide a process for the orderly resolution of these problems, e.g.: What are key problems to be resolved & in what order What alternatives exist for resolving a given problem How should mutual dependencies between the problems be handled How to resolve each individual problem most effectively in its context Help to generate & reuse software architectures Motivation Individual patterns & pattern catalogs are insufficient Software modeling methods & tools largely just illustrate what/how – not why – systems are designed

Taxonomy of Patterns & Idioms : 

Taxonomy of Patterns & Idioms

Example: Boeing Bold Stroke Product-line Architecture: 

Nav Sensors Expendable Management Data Links Mission Computer Vehicle Mgmt Expendable Radar Example: Boeing Bold Stroke Product-line Architecture

Example: Boeing Bold Stroke Product-line Architecture: 

COTS & Standards-based Middleware Infrastructure, OS, Network, & Hardware Platform Real-time CORBA middleware services VxWorks operating system VME, 1553, & Link16 PowerPC Example: Boeing Bold Stroke Product-line Architecture

Example: Boeing Bold Stroke Product-line Architecture: 

Reusable Object-Oriented Application Domain-specific Middleware Framework Configurable to variable infrastructure configurations Supports systematic reuse of mission computing functionality Example: Boeing Bold Stroke Product-line Architecture

Example: Boeing Bold Stroke Product-line Architecture: 

Product Line Component Model Configurable for product-specific functionality & execution environment Single component development policies Standard component packaging mechanisms Example: Boeing Bold Stroke Product-line Architecture

Example: Boeing Bold Stroke Product-line Architecture: 

Component Integration Model Configurable for product-specific component assembly & deployment environments Model-based component integration policies Avionics Interfaces Operator Real World Model Infrastructure Services Example: Boeing Bold Stroke Product-line Architecture

Legacy Avionics Architectures: 

Legacy Avionics Architectures Board 1 VME 1553 1: Sensors generate data Board 2 2: I/O via interrupts 3: Sensor proxies process data & pass to missions functions 4: Mission functions perform avionics operations Key System Characteristics Hard & soft real-time deadlines ~20-40 Hz Low latency & jitter between boards ~100 usecs Periodic & aperiodic processing Complex dependencies Continuous platform upgrades Avionics Mission Computing Functions Weapons targeting systems (WTS) Airframe & navigation (Nav) Sensor control (GPS, IFF, FLIR) Heads-up display (HUD) Auto-pilot (AP)

Legacy Avionics Architectures: 

Legacy Avionics Architectures Board 1 VME 1553 1: Sensors generate data Board 2 2: I/O via interrupts 3: Sensor proxies process data & pass to missions functions 4: Mission functions perform avionics operations Key System Characteristics Hard & soft real-time deadlines ~20-40 Hz Low latency & jitter between boards ~100 usecs Periodic & aperiodic processing Complex dependencies Continuous platform upgrades

Decoupling Avionics Components: 

Decoupling Avionics Components

Applying the Publisher-Subscriber Pattern to Bold Stroke: 

Applying the Publisher-Subscriber Pattern to Bold Stroke Board 1 VME 1553 1: Sensors generate data Board 2 2: I/O via interrupts 4: Event Channel pushes events to subscribers(s) 5: Subscribers perform avionics operations GPS IFF FLIR HUD Nav WTS Air Frame Publishers Subscribers push(event) push(event) Event Channel 3: Sensor publishers push events to event channel Considerations for implementing the Publisher-Subscriber pattern for mission computing applications include: Event notification model Push control vs. pull data interactions Scheduling & synchronization strategies e.g., priority-based dispatching & preemption Event dependency management e.g.,filtering & correlation mechanisms Bold Stroke uses the Publisher-Subscriber pattern to decouple sensor processing from mission computing operations Anonymous publisher & subscriber relationships Group communication Asynchrony

Ensuring Platform-neutral & Network-transparent Communication: 

Ensuring Platform-neutral & Network-transparent Communication Wrapper Facade Layers Component internal partitioning Remoting Error Lookup Requestor Object Adapter Container Facade Business Delegate Invoker Client Proxy OS abstraction request issuing request reception error notification Broker configuration component discovery request dispatching request dispatching b roker access component access component access component creation Message Publisher- Subscriber Factory Method request encapsulation publish- subscribe communication Broker

Ensuring Platform-neutral & Network-transparent Communication: 

Ensuring Platform-neutral & Network-transparent Communication operation (params) connect send_request marshal unmarshal dispatch operation (params) result marshal receive_reply unmarshal result start_up register_service assigned port Dynamics : Broker : Client Proxy : Object Adapter : Client : Server

Applying the Broker Pattern to Bold Stroke: 

Applying the Broker Pattern to Bold Stroke Board 1 VME 1553 1: Sensors generate data Board 2 2: I/O via interrupts 5: Event Channel pushes events to subscribers(s) 6: Subscribers perform avionics operations GPS IFF FLIR HUD Nav WTS Air Frame Publishers Subscribers push(event) push(event) Event Channel 4: Sensor publishers push events to event channel Bold Stroke uses the Broker pattern to shield distributed applications from environment heterogeneity, e.g., Programming languages Operating systems Networking protocols Hardware 3: Broker handles I/O via upcalls Broker A key consideration for implementing the Broker pattern for mission computing applications is QoS support e.g., latency, jitter, priority preservation, dependability, security, etc.

Slide39: 

Enables reuse of software architectures & designs Improves development team communication Convey “best practices” intuitively Transcends language-centric biases/myopia Abstracts away from many unimportant details Benefits of Patterns www.cs.wustl.edu/ ~schmidt/patterns.html

Slide40: 

Require significant tedious & error-prone human effort to handcraft pattern implementations Can be deceptively simple Leaves some important details unresolved Limitations of Patterns www.cs.wustl.edu/ ~schmidt/patterns.html

Software Design Abstractions for Concurrent & Networked Applications: 

Software Design Abstractions for Concurrent & Networked Applications Problem Distributed app & middleware functionality is subject to change since it’s often reused in unforeseen contexts, e.g., Accessed from different clients Run on different platforms Configured into different run-time contexts

Overview of Frameworks: 

Overview of Frameworks Framework Characteristics

Comparing Class Libraries, Frameworks, & Components: 

Comparing Class Libraries, Frameworks, & Components

Using Frameworks Effectively: 

Using Frameworks Effectively Observations Frameworks are powerful, but hard to develop & use effectively by application developers It’s often better to use & customize COTS frameworks than to develop in-house frameworks Components are easier for application developers to use, but aren’t as powerful or flexible as frameworks

Overview of the ACE Frameworks: 

Overview of the ACE Frameworks Features Open-source 6+ integrated frameworks 250,000+ lines of C++ 60+ person-years of effort Ported to Windows, UNIX, & real-time operating systems e.g., VxWorks, pSoS, LynxOS, Chorus, QNX Large user community www.cs.wustl.edu/~schmidt/ACE.html

The Layered Architecture of ACE: 

The Layered Architecture of ACE Features Open-source 250,000+ lines of C++ 40+ person-years of effort Ported to Win32, UNIX, & RTOSs e.g., VxWorks, pSoS, LynxOS, Chorus, QNX Large open-source user community www.cs.wustl.edu/~schmidt/ACE-users.html Commercial support by Riverace www.riverace.com/ www.cs.wustl.edu/~schmidt/ACE.html

Key Capabilities Provided by ACE: 

Key Capabilities Provided by ACE

The POSA2 Pattern Language: 

Pattern Benefits Preserve crucial design information used by applications & middleware frameworks & components Facilitate reuse of proven software designs & architectures Guide design choices for application developers The POSA2 Pattern Language

POSA2 Pattern Abstracts: 

POSA2 Pattern Abstracts Service Access & Configuration Patterns The Wrapper Facade design pattern encapsulates the functions & data provided by existing non-object-oriented APIs within more concise, robust, portable, maintainable, & cohesive object-oriented class interfaces. The Component Configurator design pattern allows an application to link & unlink its component implementations at run-time without having to modify, recompile, or statically relink the application. Component Configurator further supports the reconfiguration of components into different application processes without having to shut down & re-start running processes. The Interceptor architectural pattern allows services to be added transparently to a framework & triggered automatically when certain events occur. The Extension Interface design pattern allows multiple interfaces to be exported by a component, to prevent bloating of interfaces & breaking of client code when developers extend or modify the functionality of the component. Event Handling Patterns The Reactor architectural pattern allows event-driven applications to demultiplex & dispatch service requests that are delivered to an application from one or more clients. The Proactor architectural pattern allows event-driven applications to efficiently demultiplex & dispatch service requests triggered by the completion of asynchronous operations, to achieve the performance benefits of concurrency without incurring certain of its liabilities. The Asynchronous Completion Token design pattern allows an application to demultiplex & process efficiently the responses of asynchronous operations it invokes on services. The Acceptor-Connector design pattern decouples the connection & initialization of cooperating peer services in a networked system from the processing performed by the peer services after they are connected & initialized.

POSA2 Pattern Abstracts (cont’d): 

POSA2 Pattern Abstracts (cont’d) Synchronization Patterns The Scoped Locking C++ idiom ensures that a lock is acquired when control enters a scope & released automatically when control leaves the scope, regardless of the return path from the scope. The Strategized Locking design pattern parameterizes synchronization mechanisms that protect a component’s critical sections from concurrent access. The Thread-Safe Interface design pattern minimizes locking overhead & ensures that intra-component method calls do not incur ‘self-deadlock’ by trying to reacquire a lock that is held by the component already. The Double-Checked Locking Optimization design pattern reduces contention & synchronization overhead whenever critical sections of code must acquire locks in a thread-safe manner just once during program execution. Concurrency Patterns The Active Object design pattern decouples method execution from method invocation to enhance concurrency & simplify synchronized access to objects that reside in their own threads of control. The Monitor Object design pattern synchronizes concurrent method execution to ensure that only one method at a time runs within an object. It also allows an object’s methods to cooperatively schedule their execution sequences. The Half-Sync/Half-Async architectural pattern decouples asynchronous & synchronous service processing in concurrent systems, to simplify programming without unduly reducing performance. The pattern introduces two intercommunicating layers, one for asynchronous & one for synchronous service processing. The Leader/Followers architectural pattern provides an efficient concurrency model where multiple threads take turns sharing a set of event sources in order to detect, demultiplex, dispatch, & process service requests that occur on the event sources. The Thread-Specific Storage design pattern allows multiple threads to use one ‘logically global’ access point to retrieve an object that is local to a thread, without incurring locking overhead on each object access.

Implementing the Broker Pattern for Bold Stroke Avionics: 

Implementing the Broker Pattern for Bold Stroke Avionics CORBA is a distribution middleware standard Real-time CORBA adds QoS to classic CORBA to control: www.omg.org 3. Memory Resources These capabilities address some (but by no means all) important DRE application development & QoS-enforcement challenges 2. Communication Resources 1. Processor Resources Request Buffering

Example of Applying Patterns & Frameworks to Middleware:Real-time CORBA & The ACE ORB (TAO): 

Example of Applying Patterns & Frameworks to Middleware: Real-time CORBA & The ACE ORB (TAO) TAO Features Open-source 500,000+ lines of C++ ACE/patterns-based 50+ person-years of effort Ported to UNIX, Win32, MVS, & many RT & embedded OSs e.g., VxWorks, LynxOS, Chorus, QNX www.cs.wustl.edu/~schmidt/TAO.html Large open-source user community www.cs.wustl.edu/~schmidt/TAO-users.html Commercially supported www.theaceorb.com, www.remedy.nl, www.prismtechnologies.com

Key Patterns Used in TAO: 

Key Patterns Used in TAO www.cs.wustl.edu/~schmidt/PDF/ORB-patterns.pdf Wrapper facades enhance portability Proxies & adapters simplify client & server applications, respectively Component Configurator dynamically configures Factories Factories produce Strategies Strategies implement interchangeable policies Concurrency strategies use Reactor & Leader/Followers Acceptor-Connector decouples connection management from request processing Managers optimize request demultiplexing

Enhancing ORB Flexibility w/the Strategy Pattern : 

Enhancing ORB Flexibility w/the Strategy Pattern

Consolidating Strategies with the Abstract Factory Pattern: 

Consolidating Strategies with the Abstract Factory Pattern

Dynamically Configuring Factories w/the Component Configurator Pattern: 

Dynamically Configuring Factories w/the Component Configurator Pattern

ACE Frameworks Used in TAO: 

ACE Frameworks Used in TAO Reactor drives the ORB event loop Implements the Reactor & Leader/Followers patterns Acceptor-Connector decouples passive/active connection roles from GIOP request processing Implements the Acceptor-Connector & Strategy patterns Service Configurator dynamically configures ORB strategies Implements the Component Configurator & Abstract Factory patterns www.cs.wustl.edu/~schmidt/PDF/ICSE-03.pdf

Summary of Pattern, Framework, & Middleware Synergies: 

Summary of Pattern, Framework, & Middleware Synergies These technologies codify expertise of skilled researchers & developers There are now powerful feedback loops advancing these technologies

Example:Electronic Medical Imaging Systems: 

Example: Electronic Medical Imaging Systems Goal Route, manage, & manipulate electronic medical images robustly, efficiently, & securely thoughout a distributed health care environment Modalities e.g., MRI, CT, CR, Ultrasound, etc.

Example:Electronic Medical Imaging Systems: 

Example: Electronic Medical Imaging Systems System Characteristics Large volume of “blob” data e.g.,10 to 40 Mbps “Lossy” compression isn’t viable due to liability concerns Diverse QoS requirements, e.g., Sync & async communication Event- & method-based invocation Streaming communication Prioritization of requests & streams Distributed resource management

Example:Electronic Medical Imaging Systems: 

Example: Electronic Medical Imaging Systems

Image Acquisition Scenario: 

Modalities e.g., MRI, CT, CR, Ultrasound, etc. Image Acquisition Scenario Diagnostic & Clinical Workstations Key Tasks Image location & routing Image delivery Image Database

Slide63: 

Applying Patterns to Resolve Key Distributed System Design Challenges Patterns help resolve the following common design challenges:

Separating Concerns Between Tiers: 

Separating Concerns Between Tiers Context Distributed systems are now common due to the advent of The global Internet Ubiquitous mobile & embedded devices Problem It’s hard to build distributed systems due to the complexity associated with many capabilities at many levels of abstraction

Applying the Layers Pattern to Image Acquisition: 

Applying the Layers Pattern to Image Acquisition Diagnostic & clinical workstations are presentation tier entities that: Typically represent sophisticated GUI elements Share the same address space with their clients Their clients are containers that provide all the resources Exchange messages with the middle tier components

Pros & Cons of the Layers Pattern: 

Pros & Cons of the Layers Pattern This pattern has four benefits: Reuse of layers If an individual layer embodies a well-defined abstraction & has a well-defined & documented interface, the layer can be reused in multiple contexts Support for standardization Clearly-defined & commonly-accepted levels of abstraction enable the development of standardized tasks & interfaces Dependencies are localized Standardized interfaces between layers usually confine the effect of code changes to the layer that is changed Exchangeability Individual layer implementations can be replaced by semantically-equivalent implementations without undue effort This pattern also has liabilities: Cascades of changing behavior If layer interfaces & semantics aren’t abstracted properly then changes can ripple when behavior of a layer is modified Higher overhead A layered architecture can be less efficient than a monolithic architecture Unnecessary work If some services performed by lower layers perform excessive or duplicate work not actually required by the higher layer, performance can suffer Difficulty of establishing the correct granularity of layers It’s important to avoid too many & too few layers

Overview of Distributed Object Computing Communication Mechanisms: 

Overview of Distributed Object Computing Communication Mechanisms Solution DOC middleware provides multiple types of communication mechanisms: Collocated client/server (i.e., native function call) Synchronous & asynchronous RPC/IPC Group communication & events Data streaming Problem A single communication mechanism does not fit all uses! Context In multi-tier systems both the tiers & the components within the tiers must be connected via communication mechanisms We now explore various distribution infrastructure (P4) patterns that applications can apply to leverage these communication mechanisms

Core Distribution Infrastructure Patterns: 

Core Distribution Infrastructure Patterns Broker makes invocations on remote component objects look & act as much as possible like invocations on component objects in the same address space as their clients Messaging & Publisher-Subscriber are most appropriate for integration scenarios where multiple, independently developed & self-contained services or applications must collaborate & form a coherent software system

Improving Type-safety & Performance (1/2): 

Improving Type-safety & Performance (1/2) Context The configuration of components in distributed systems is often subject to change as requirements evolve Problems Low-level message passing (e.g., using sockets) is error-prone & fraught with accidental complexity Remote components should look like local components from an application perspective i.e., ideally clients & servers should be oblivious to communication mechanisms & locations

Improving Type-safety & Performance (2/2): 

Improving Type-safety & Performance (2/2) A Service implements the object, which is not accessible directly A Proxy represents the Service & ensures the correct access to it Proxy offers same interface as Service Clients use the Proxy to access the Service Client 1 1

Applying the Proxy Pattern to Image Acquisition: 

Applying the Proxy Pattern to Image Acquisition Client 1 1 We can apply the Proxy pattern to provide a strongly-typed interface to initiate & coordinate the downloading of images from an image database

Pros & Cons of the Proxy Pattern: 

Pros & Cons of the Proxy Pattern This pattern provides three benefits: Decoupling clients from the location of server components Clients are not affected by migration of servers or changes in the networking infrastructure since location information & addressing functionality is managed by a proxy Separation of housekeeping & functionality A proxy relieves clients from burdens that do not inherently belong to the task the client performs Potential for time & space optimizations Proxy implementations can be loaded “on-demand” & can also be used to cache values to avoid redundant remote calls Proxies can also be optimized to improve both type-safety & performance This pattern has two liabilities: Potential overkill via sophisticated strategies If proxies include overly sophisticated functionality, such as caching or replica management, they many introduce overhead that defeats their intended purpose Higher overhead due to indirection Proxies introduce an additional layer of indirection that can be excessive if the proxy implementation is inefficient

Enabling Client Extensibility (1/2): 

Enabling Client Extensibility (1/2) Context Object models define how components import & export functionality e.g., UML class diagrams specify well-defined OO interfaces Class X operation1() operation2() operation3() operationn()

Enabling Client Extensibility (2/2): 

Enabling Client Extensibility (2/2)

Extension Interface Pattern Dynamics: 

Extension Interface Pattern Dynamics : Client Start_client : Factory createInstance(Ext.Intf. 1) new Ref. To Extension1 create create service_1 queryInterface(Extension Interface 2) Ref. To Extension2 Note how each extension interface can serve as a “factory” to return object reference to other extension interfaces supported by a component : Component : Extension 1 : Extension 2 service_2 A common use of the Extension Interface pattern is to support component versioning Off-loads versioning protocol to client…

Pros & Cons of the Extension Interface Pattern: 

Pros & Cons of the Extension Interface Pattern This pattern has five benefits: Separation of concerns Interfaces are strictly decoupled from implementations Exchangeability of components Component implementations can evolve independently from clients that access them Extensibility through interfaces Clients only access components via their interfaces, which reduces coupling to representation & implementation details Prevention of interface bloating Interfaces need not contain all possible methods, just the ones associated with a particular capability No subclassing required Delegation—rather than inheritance—is used to customize components This pattern also has liabilities: Higher overhead due to indirection Clients must incur the overhead of several round-trips to obtain the appropriate object reference from a server component Complexity & cost for development & deployment This pattern off-loads the responsibility for determining the appropriate interface from the component designer to the client applications

Ensuring Platform-neutral & Network-transparent OO Communication (1/2): 

Ensuring Platform-neutral & Network-transparent OO Communication (1/2) Problem A middleware architecture needs to: Support remote method invocation Provide location transparency Detect & recover from faults Allow the addition, exchange, or remove of services dynamically Hide system details from developers Context The Proxy & Extension Interface patterns aren’t sufficient since they don’t address how Remote components are located Connections are established Messages are exchanged across a network etc. Client 1 1

Ensuring Platform-neutral & Network-transparent OO Communication (2/2): 

Ensuring Platform-neutral & Network-transparent OO Communication (2/2) Wrapper Facade Layers Component internal partitioning Remoting Error Lookup Requestor Object Adapter Container Facade Business Delegate Invoker Client Proxy OS abstraction request issuing request reception error notification Broker configuration component discovery request dispatching request dispatching b roker access component access component access component creation Message Publisher- Subscriber Factory Method request encapsulation publish- subscribe communication Broker

Broker Pattern Dynamics: 

Broker Pattern Dynamics method (proxy) locate_server server port send_request marshal unmarshal dispatch method (impl.) result marshal receive_reply unmarshal result start_up register_service : Broker : Client Proxy : Server Proxy : Client : Server assigned port Broker middleware generates the necessary client & server proxies from higher level interface definitions Proxy Generator

Applying the Broker Pattern to Image Acquisition : 

Applying the Broker Pattern to Image Acquisition Common Broker pattern implementations CORBA COM+ Java RMI Med-specific Comm (MSC) Brokers define interfaces… … not implementations Brokers simplify development of distributed applications by automating Object location Connection management Memory management Parameter (de)marshaling Event & request demuxing Error handling Object/server activation Concurrency Brokers help shield distributed applications from environment heterogeneity e.g., programming languages, operating systems, networking protocols, hardware, etc.

Pros & Cons of the Broker Pattern: 

Pros & Cons of the Broker Pattern This pattern has five benefits: Portability enhancements A broker hides OS & network system details from clients & servers by using indirection & abstraction layers, such as APIs, proxies, adapters, & bridges Interoperability with other brokers Different brokers may interoperate if they understand a common protocol for exchanging messages Reusability of services When building new applications, brokers enable application functionality to reuse existing services Location transparency A broker is responsible for locating servers, so clients need not know where servers are located Changeability & extensibility of components If server implementations change without affecting interfaces clients should not be affected This pattern also has liabilities: Higher overhead Applications using brokers may be slower than applications written manually Potentially less reliable Compared with non-distributed software applications, distributed broker systems may incur lower fault tolerance Testing & debugging may be harder Testing & debugging of distributed systems is tedious because of all the components involved

Supporting Async Communication (1/2): 

Supporting Async Communication (1/2) Context Some clients want to send requests, continue their work, & receive the results at some later point in time Problem Broker implementations based on conventional RPC semantics often just support blocking operations i.e., clients must wait until twoway invocations return Unfortunately, this design can reduce scalability & complicate certain use-cases

Supporting Async Communication (1/2): 

Supporting Async Communication (1/2) Introduce intermediary queue(s) between clients & servers: A queue is used to store messages A queue can cooperate with other queues to route messages Messages are sent from sender to receiver A client sends a message, which is queued & then forwarded to a message processor on a server that receives & executes them A Message API is provided for clients & servers to send/receive messages

Messaging Pattern Dynamics: 

Messaging Pattern Dynamics : Client : Message create store Message forward Message : Message Processor create receive receive Message Message exec send Message : Message API : Local Queue : Remote Queue : Message API Reply store forward Reply recv Reply send Reply recv Other processing

Applying the Messaging Pattern to Image Acquisition: 

Applying the Messaging Pattern to Image Acquisition We can apply the Messaging pattern to Queue up image request messages remotely without blocking the diagnostic/clinical workstation clients Execute the requests at a later point & return the results to the client This design also enables other, more advanced capabilities, e.g., Multi-hop store & forward persistence QoS-driven routing, where requests can be delivered to the “best” image database depending on context

Pros & Cons of the Messaging Pattern: 

This pattern provides three benefits: Enhances concurrency by transparently leveraging available parallelism Messages can be executed remotely on servers while clients perform other processing Simplifies synchronized access to a shared object that resides in its own thread of control Since messages are processed serially by a message processor target objects often need not be concerned with synchronization mechanisms Message execution order can differ from message invocation order This allows reprioritizing of messages to enhance quality of service Messages can be “batched” & sent wholesale to enhance throughout This pattern also has some liabilities: Message execution order can differ from message invocation order As a result, clients must be careful not to rely on ordering dependencies Lack of type-safety Clients & servers are responsible for formatting & passing messages Complicated debugging As with all distributed systems, debugging & testing is complex Pros & Cons of the Messaging Pattern

Supporting OO Async Communication (1/2): 

Supporting OO Async Communication (1/2) Context Some clients want to invoke remote operations, continue their work, & retrieve the results at a later point in time Problem Using the explicit message-passing API of the Messaging pattern can reduce type-safety & performance Similar to motivation for Proxy pattern...

Supporting OO Async Communication (2/2): 

Supporting OO Async Communication (2/2) A proxy provides an interface that allows clients to access methods of an object A concrete method request is created for every method invoked on the proxy A scheduler receives the method requests & dispatches them on the servant when they become runnable An activation list maintains pending method requests A servant implements the methods A future allows clients to access the results of a method call on the proxy

Active Object Pattern Dynamics: 

A client invokes a method on the proxy The proxy returns a future to the client, & creates a method request, which it passes to the scheduler The scheduler enqueues the method request into the activation list (not shown here) When the method request becomes runnable, the scheduler dequeues it from the activation list (not shown here) & executes it in a different thread than the client The method request executes the method on the servant & writes results, if any, to the future Clients obtain the method’s results via the future Active Object Pattern Dynamics Clients can obtain result from futures via blocking, polling, or callbacks

Applying the Active Object Pattern to Image Acquisition: 

Applying the Active Object Pattern to Image Acquisition OO developers often prefer method-oriented request/response semantics to message-oriented semantics The Active Object pattern supports this preference via strongly-typed async method APIs: Several types of parameters can be passed: Requests contain in/inout arguments Results carry out/inout arguments & results Callback object or poller object can be used to retrieve results get_image() future results

Pros & Cons of the Active Object Pattern: 

This pattern provides four benefits: Enhanced type-safety Cf. async forwarder/receiver message passing Enhances concurrency & simplifies synchronized complexity Concurrency is enhanced by allowing client threads & asynchronous method executions to run simultaneously Synchronization complexity is simplified by using a scheduler that evaluates synchronization constraints to serialized access to servants Transparent leveraging of available parallelism Multiple active object methods can execute in parallel if supported by the OS/hardware Method execution order can differ from method invocation order Methods invoked asynchronous are executed according to the synchronization constraints defined by their guards & by scheduling policies Methods can be “batched” & sent wholesale to enhance throughput This pattern also has some liabilities: Higher overhead Depending on how an active object’s scheduler is implemented, context switching, synchronization, & data movement overhead may occur when scheduling & executing active object invocations Complicated debugging It is hard to debug programs that use the Active Object pattern due to the concurrency & non-determinism of the various active object schedulers & the underlying OS thread scheduler Pros & Cons of the Active Object Pattern

Decoupling Suppliers & Consumers (1/2): 

Decoupling Suppliers & Consumers (1/2) Problem Having each client call a specific server is inefficient & non-scalable A “polling” strategy leads to performance bottlenecks Work lists could be spread across different servers More than one client may be interested in work list content Context In large-scale electronic medical imaging systems, radiologists may share “work lists” of patient images to balance workloads effectively

Decoupling Suppliers & Consumers (2/2): 

Decoupling Suppliers & Consumers (2/2) Decouple suppliers (publishers) & consumers (subscribers) of events: An Event Channel stores/forwards events Publishers create events & store them in a queue maintained by the Event Channel Consumers register with event queues, from which they retrieve events Events are used to transmit state change info from publishers to consumers For event transmission push-models & pull-models are possible Filters can filter events for consumers Solution Apply the Publisher-Subscriber architectural pattern (P1) to decouple image suppliers from consumers

Publisher-Subscriber Pattern Dynamics: 

Publisher-Subscriber Pattern Dynamics The Publisher-Subscriber pattern helps keep the state of cooperating components synchronized To achieve this it enables one-way propagation of changes: one publisher notifies any number of subscribers about changes to its state attachSubscriber produce pushEvent event event pushEvent detachSubscriber : Event : Subscriber : Event Channel : Publisher Key design considerations for the Publisher-Subscriber pattern include: Push vs. pull interaction models Control vs. data event notification models Multicast vs. unicast communication models Persistence vs. transient queueing models

Applying the Publisher-Subscriber Pattern to Image Acquisition: 

Applying the Publisher-Subscriber Pattern to Image Acquisition * creates receives Event Channel attachPublisher detachPublisher attachSubscriber detachSubscriber pushEvent Radiologists can subscribe to an event channel to receive notifications produced when modalities publish events indicating the arrival of new images & studies This design enables a group of distributed radiologists to collaborate effectively in a networked environment

Pros & Cons of the Publisher-Subscriber Pattern: 

Pros & Cons of the Publisher-Subscriber Pattern This pattern has two benefits: Decouples consumers & producers of events All an event channel knows is that it has a list of consumers, each conforming to the simple interface of the Subscriber class The coupling between the publishers & subscribers is therefore abstract, anonymous, & minimal n:m communication models are supported Unlike an ordinary sync/async request/response invocation, the notification that a publisher sends needn’t designate its receiver, which enables a broader range of communication topologies, including multicast & broadcast There are also liabilities: Must be careful with potential update cascades Since subscribers have no knowledge of each other’s presence, applications may not recognize the ultimate cost of publishing events through an event channel A seemingly innocuous operation on the subject may therefore cause a cascade of updates to observers & their dependent objects Performance degradation relative to point-to-point request/response interactions

Locating & Creating Components Scalably (1/2): 

Locating & Creating Components Scalably (1/2) Context Our electronic medical imaging system contains many components distributed in a network Problem How to create new components and/or find existing ones Simple solutions appropriate for stand-alone applications don’t scale “Obvious” distributed solutions also don’t scale

Locating & Creating Components Scalably (2/2): 

Locating & Creating Components Scalably (2/2) An Abstract Home declares an interface for operations that find and/or create abstract instances of components Concrete Homes implements the abstract Home interface to find specific instances and/or create new ones Abstract Comp declares interface for a specific type of component class Concrete Comp define instances A Primary Key is associated with a component

Factory/Finder Pattern Dynamics: 

Factory/Finder Pattern Dynamics The Factory/Finder pattern is supported by distributed component models e.g., EJB, COM+, & the CCM

Applying the Factory/Finder Pattern to Image Acquisition: 

Applying the Factory/Finder Pattern to Image Acquisition We can apply the Factory/Finder pattern to create/locate image transfer components for images needed by radiologists If a suitable component already exists the component home will use it, otherwise, it will create a new component

Pros & Cons of the Factory/Finder Pattern: 

Pros & Cons of the Factory/Finder Pattern This pattern has three benefits: Separation of concerns Finding/creating individual components is decoupled from locating the factories that find/create these components Improved scalability e.g., general-purpose directory mechanisms need not manage the creation & location of large amounts of finer-grained components whose lifetimes may be short Customized capabilities The location/creation mechanism can be specialized to support key capabilities that are unique for various types of components This pattern also has some liabilities: Overhead due to indirection Clients must incur the overhead of several round-trips to obtain the appropriate object reference Complexity & cost for development & deployment There are more steps involved in obtaining object references, which can complicate client programming

Extending Components Transparently: 

Context Component developers may not know a priori the context in which their components will execute Thus, containers are introduced to: Shield clients & components from the details of the underlying middleware, services, network, & OS Manage the lifecycle of components & notify them about lifecycle events e.g., activation, passivation, & transaction progress Provide components with uniform access to middleware infrastructure services e.g., transactions, security, & persistence Register & deploy components Declarative Programming Transaction Security Resources ... Transaction Security Resources ... ... Imperative Programming Client Client Extending Components Transparently

Extending Components Transparently (cont‘d): 

Extending Components Transparently (cont‘d) Problem Developers should be able to specify declaratively what type of execution environment components need e.g., in configuration files or databases Containers must be able to transparently provide the right execution environment e.g., by creating a new transaction or new servant when required Framework represents the concrete framework to which we attach interceptors Concrete Interceptors implement the event handler for the system-specific events they have subscribed for Context contains information about the event & allows modification of system behavior after interceptor completion The Dispatcher allows applications to register & remove interceptors with the framework & to delegate events to interceptors

Interceptor Pattern Dynamics: 

Interceptor Pattern Dynamics Interceptor are a “meta-programming mechanism,” along with Smart proxies Pluggable protocols Gateways/bridges Interface repositories These mechanisms provide building-blocks to handle (often unanticipated) variation translucently & reflectively More information on meta-programming mechanisms can be found at Interception is commonly used to handle security & transactions transparently from the perspective of a component implementation It can also enable performance enhancement strategies e.g., just-in-time activation, object pooling, load balancing, & caching : Application : Framework : Interceptor create run_event_loop attach interceptor Place interceptor in internal interceptor map event Look for registered interceptors : Context create context handle_event : Dispatcher delegate www.cs.wustl.edu/~schmidt/PDF/IEEE.pdf

Applying the Interceptor Pattern to Image Acquisition: 

Applying the Interceptor Pattern to Image Acquisition A container provides generic interfaces to a component that it can use to access container functionality e.g., transaction control, persistence, security,load balancing etc. A container intercepts all incoming requests from clients, e.g., It can read the component’s requirements from a XML configuration file It can then do some pre-processing before actually delegating the request to the component A component provides interfaces that its container invokes automatically when particular events occur e.g., activation or passivation Interceptors are used for many other middleware tasks, as well

Pros & Cons of the Interceptor Pattern: 

Pros & Cons of the Interceptor Pattern This pattern has five benefits: Extensibility & flexibility Interceptors allow an application to evolve without breaking existing APIs & components Separation of concerns Interceptors decouple the “functional” path from the “meta” path Support for monitoring & control of frameworks e.g., generic logging mechanisms can be used to unobtrusively track application behavior Layer symmetry Interceptors can perform transformations on a client-side whose inverse are performed on the server-side Reusability Interceptors can be reused for various general-purpose behaviors This pattern also has liabilities: Complex design issues Determining interceptor APIs & semantics is non-trivial Malicious or erroneous interceptors Mis-behaving interceptors can wreak havoc on application stability Potential interception cascades Interceptors can result in infinite recursion

Minimizing Resource Utilization (1/2): 

Minimizing Resource Utilization (1/2) Problem It may not feasible to have all image server implementations running all the time since this ties up end-system resources unnecessarily Context Image servers are simply one of many services running throughout a distributed electronic medical image system

Minimizing Resource Utilization (2/2): 

Minimizing Resource Utilization (2/2) www.cs.wustl.edu/~schmidt/PDF/Activator.pdf When incoming requests arrive, the Activator looks up whether a target object is already active & if the object is not running it activates the Service Execution Context The Activation Table stores associations between services & their physical location The Client uses the Activator to get service access A Service implements a specific type of functionality that it provides to clients

Activator Pattern Dynamics: 

Activator Pattern Dynamics

Applying the Activator Pattern to Image Acquisition: 

Applying the Activator Pattern to Image Acquisition Activator createService() findService() activate() deactivate() addService() remService() Activation Table lookup() insert() delete() getService useService() Server (ringil:5500) 2.1 start The Activator pattern is available in various COTS technologies: UNIX Inetd “super server” CORBA Implementation Repository We can use the Activator pattern to launch image transfer servers on-demand iiop://ringil:5500/poa_name/object_name

Pros & Cons of the Activator Pattern: 

Pros & Cons of the Activator Pattern This pattern has three benefits: More effective resource utilization Servers can be spawned “on-demand,” thereby minimizing resource utilization until clients actually require them Uniformity By imposing a uniform activation interface to spawn & control servers Modularity, testability, & reusability Application modularity & reusability is improved by decoupling server implementations from the manner in which the servers are activated This pattern also has liabilities: Lack of determinism & ordering dependencies This pattern makes it hard to determine or analyze the behavior of an application until its components are activated at run-time Reduced security or reliability An application that uses the Activator pattern may be less secure or reliable than an equivalent statically-configured application Increased run-time overhead & infrastructure complexity By adding levels of abstraction & indirection when activating & executing components

Enhancing Server (Re)Configurability (1/2): 

Enhancing Server (Re)Configurability (1/2) Certain factors are static, such as the number of available CPUs & operating system support for asynchronous I/O Other factors are dynamic, such as system workload Context The implementation of certain image server components depends on a variety of factors: Problem Prematurely committing to a particular image server component configuration is inflexible & inefficient: No single image server configuration is optimal for all use cases Certain design decisions cannot be made efficiently until run-time

Enhancing Server (Re)Configurability (2/2): 

Enhancing Server (Re)Configurability (2/2) This pattern allows an application to link & unlink its component implementations at run-time Thus, new & enhanced services can be added without having to modify, recompile, statically relink, or shut down & restart a running application

Component Configurator Pattern Dynamics: 

Component Configurator Pattern Dynamics run_component() run_component() fini() remove() remove() fini() Comp. A Concrete Comp. B Concrete Comp. A Concrete Comp. B Component initialization & dynamic linking Component processing Component termination & dynamic unlinking

Applying the Component Configurator Pattern to Image Acquisition: 

Applying the Component Configurator Pattern to Image Acquisition <<contains>> components * Component Configurator Component Repository LRU File Cache LFU File Cache Component init() fini() suspend() resume() info() For example, an image server can apply the Component Configurator pattern to configure various Cached Virtual Filesystem strategies e.g., least-recently used (LRU) or least-frequently used (LFU) Image servers can use the Component Configurator pattern to dynamically optimize, control, & reconfigure the behavior of its components at installation-time or during run-time

Reconfiguring an Image Server: 

Reconfiguring an Image Server Image servers can also be reconfigured dynamically to support new components & new component implementations

Pros & Cons of the Component Configurator Pattern: 

Pros & Cons of the Component Configurator Pattern This pattern offers four benefits: Uniformity By imposing a uniform configuration & control interface to manage components Centralized administration By grouping one or more components into a single administrative unit that simplifies development by centralizing common component initialization & termination activities Modularity, testability, & reusability Application modularity & reusability is improved by decoupling component implementations from the manner in which the components are configured into processes Configuration dynamism & control By enabling a component to be dynamically reconfigured without modifying, recompiling, statically relinking existing code & without restarting the component or other active components with which it is collocated This pattern also incurs liabilities: Lack of determinism & ordering dependencies This pattern makes it hard to determine or analyze the behavior of an application until its components are configured at run-time Reduced security or reliability An application that uses the Component Configurator pattern may be less secure or reliable than an equivalent statically-configured application Increased run-time overhead & infrastructure complexity By adding levels of abstraction & indirection when executing components Overly narrow common interfaces The initialization or termination of a component may be too complicated or too tightly coupled with its context to be performed in a uniform manner

Example:High-performance Content Delivery Servers: 

Example: High-performance Content Delivery Servers Support many content delivery server design alternatives seamlessly e.g., different concurrency & event models Design is guided by patterns to leverage time-proven solutions Key Solution Characteristics Implementation based on COTS framework components to reduce effort & amortize prior work Open-source to control costs & to leverage technology advances Key System Characteristics Robust implementation e.g., stop malicious clients Extensible to other protocols e.g., HTTP 1.1, IIOP, DICOM Leverage advanced multi-processor hardware & software Goal Download content scalably & efficiently e.g., images & other multi-media content types

JAWS Content Server Framework: 

JAWS Content Server Framework Key Sources of Variation Concurrency models e.g.,thread pool vs. thread-per-connection Event demultiplexing models e.g.,sync vs. async File caching models e.g.,LRU vs. LFU Content delivery protocols e.g.,HTTP 1.0+1.1 vs. IIOP Operating system APIs e.g., Windows, UNIX, RTOS Event Dispatcher Accepts client connection request events, receives HTTP GET requests, & coordinates JAWS’s event demultiplexing strategy with its concurrency strategy Protocol Handler Performs parsing & protocol processing of HTTP request events. Cached Virtual Filesystem Improves Web server performance by reducing the overhead of file system accesses when processing HTTP GET requests

Applying Patterns to Resolve Key JAWS Design Challenges: 

Applying Patterns to Resolve Key JAWS Design Challenges Patterns help resolve the following common design challenges:

Encapsulating Low-level OS APIs (1/2): 

Encapsulating Low-level OS APIs (1/2) Problem The diversity of hardware & operating systems makes it hard to build portable & robust Web server software Programming directly to low-level OS APIs is tedious, error-prone, & non-portable Context A Web server must manage a variety of OS services, including processes, threads, Socket connections, virtual memory, & files OS platforms provide low-level APIs written in C to access these services

Encapsulating Low-level OS APIs (2/2): 

Encapsulating Low-level OS APIs (2/2) This pattern encapsulates data & functions provided by existing non-OO APIs within more concise, robust, portable, maintainable, & cohesive OO class interfaces

Applying the Wrapper Façade Pattern in JAWS: 

Applying the Wrapper Façade Pattern in JAWS JAWS uses the wrapper facades defined by ACE to ensure its framework components can run on many OS platforms e.g., Windows, UNIX, & many real-time operating systems

Pros & Cons of the Wrapper Façade Pattern: 

Pros & Cons of the Wrapper Façade Pattern This pattern provides three benefits: Concise, cohesive, & robust higher-level object-oriented programming interfaces These interfaces reduce the tedium & increase the type-safety of developing applications, which descreases certain types of programming errors Portability & maintainability Wrapper facades can shield application developers from non-portable aspects of lower-level APIs Modularity, reusability & configurability This pattern creates cohesive & reusable class components that can be ‘plugged’ into other components in a wholesale fashion, using object-oriented language features like inheritance & parameterized types This pattern can incur liabilities: Loss of functionality Whenever an abstraction is layered on top of an existing abstraction it is possible to lose functionality Performance degradation This pattern can degrade performance if several forwarding function calls are made per method Programming language & compiler limitations It may be hard to define wrapper facades for certain languages due to a lack of language support or limitations with compilers

Decoupling Event Demuxing, Connection Management, & Protocol Processing (1/2): 

Decoupling Event Demuxing, Connection Management, & Protocol Processing (1/2) Context Thus, changes to event-demuxing & connection code affects server protocol code directly & may yield subtle bugs, e.g., when porting to use TLI or WaitForMultipleObjects() select (width, &read_handles, 0, 0, 0); if (FD_ISSET (acceptor, &ready_handles)) { int h; do { h = accept (acceptor, 0, 0); char buf[BUFSIZ]; for (ssize_t i; (i = read (h, buf, BUFSIZ)) > 0; ) write (1, buf, i); } while (h != -1); Problem Developers often couple event-demuxing & connection code with protocol-handling code This code cannot then be reused directly by other protocols or by other middleware & applications

Decoupling Event Demuxing, Connection Management, & Protocol Processing (2/2): 

Decoupling Event Demuxing, Connection Management, & Protocol Processing (2/2) Handle owns dispatches * notifies * * handle set Reactor handle_events() register_handler() remove_handler() Event Handler handle_event () get_handle() Connector Synchronous Event Demuxer select () <<uses>> Acceptor Service Handler

The Reactor Pattern: 

The Reactor Pattern The Reactor architectural pattern allows event-driven applications to demultiplex & dispatch service requests that are delivered to an application from one or more clients

Reactor Pattern Dynamics: 

Reactor Pattern Dynamics : Main Program : Concrete Event Handler : Reactor : Synchronous Event Demultiplexer register_handler() get_handle() handle_events() select() handle_event() Handle Handles Handles Con. Event Handler Events service() event Observations Note inversion of control Also note how long-running event handlers can degrade the QoS since callbacks steal the reactor’s thread! Initialize phase Event handling phase

The Acceptor-Connector Pattern: 

The Acceptor-Connector Pattern The Acceptor-Connector design pattern decouples the connection & initialization of cooperating peer services in a networked system from the processing performed by the peer services after being connected & initialized

Acceptor Dynamics: 

Acceptor Dynamics ACCEPT_ EVENT Handle1 Acceptor : Handle2 Handle2 Handle2 Passive-mode endpoint initialize phase Service handler initialize phase Service processing phase The Acceptor ensures that passive-mode transport endpoints aren’t used to read/write data accidentally And vice versa for data transport endpoints… There is typically one Acceptor factory per-service/per-port Additional demuxing can be done at higher layers, a la CORBA

Synchronous Connector Dynamics: 

Synchronous Connector Dynamics Motivation for Synchrony Sync connection initiation phase Service handler initialize phase Service processing phase If the services must be initialized in a fixed order & the client can’t perform useful work until all connections are established If connection latency is negligible e.g., connecting with a server on the same host via a ‘loopback’ device If multiple threads of control are available & it is efficient to use a thread-per-connection to connect each service handler synchronously

Asynchronous Connector Dynamics: 

Asynchronous Connector Dynamics Motivation for Asynchrony Async connection initiation phase Service handler initialize phase Service processing phase If client is initializing many peers that can be connected in an arbitrary order If client is establishing connections over high latency links If client is a single-threaded application

Applying the Reactor & Acceptor-Connector Patterns in JAWS: 

Applying the Reactor & Acceptor-Connector Patterns in JAWS handle_event () get_handle() handle_event () get_handle() owns dispatches * notifies * * handle set ACE_Reactor handle_events() register_handler() remove_handler() ACE_Event_Handler handle_event () get_handle() HTTP Acceptor HTTP Handler Synchronous Event Demuxer select () <<uses>> The Reactor architectural pattern decouples: JAWS generic synchronous event demultiplexing & dispatching logic from The HTTP protocol processing it performs in response to events ACE_Handle

Reactive Connection Management & Data Transfer in JAWS: 

Reactive Connection Management & Data Transfer in JAWS

Pros & Cons of the Reactor Pattern: 

Pros & Cons of the Reactor Pattern This pattern offers four benefits: Separation of concerns This pattern decouples application-independent demuxing & dispatching mechanisms from application-specific hook method functionality Modularity, reusability, & configurability This pattern separates event-driven application functionality into several components, which enables the configuration of event handler components that are loosely integrated via a reactor Portability By decoupling the reactor’s interface from the lower-level OS synchronous event demuxing functions used in its implementation, the Reactor pattern improves portability Coarse-grained concurrency control This pattern serializes the invocation of event handlers at the level of event demuxing & dispatching within an application process or thread This pattern can incur liabilities: Restricted applicability This pattern can be applied efficiently only if the OS supports synchronous event demuxing on handle sets Non-pre-emptive In a single-threaded application, concrete event handlers that borrow the thread of their reactor can run to completion & prevent the reactor from dispatching other event handlers Complexity of debugging & testing It is hard to debug applications structured using this pattern due to its inverted flow of control, which oscillates between the framework infrastructure & the method call-backs on application-specific event handlers

Pros & Cons of Acceptor-Connector Pattern: 

Pros & Cons of Acceptor-Connector Pattern This pattern provides three benefits: Reusability, portability, & extensibility This pattern decouples mechanisms for connecting & initializing service handlers from the service processing performed after service handlers are connected & initialized Robustness This pattern strongly decouples the service handler from the acceptor, which ensures that a passive-mode transport endpoint can’t be used to read or write data accidentally Efficiency This pattern can establish connections actively with many hosts asynchronously & efficiently over long-latency wide area networks Asynchrony is important in this situation because a large networked system may have hundreds or thousands of host that must be connected This pattern also has liabilities: Additional indirection The Acceptor-Connector pattern can incur additional indirection compared to using the underlying network programming interfaces directly Additional complexity The Acceptor-Connector pattern may add unnecessary complexity for simple client applications that connect with only one server & perform one service using a single network programming interface

Overview of Concurrency & Threading: 

Overview of Concurrency & Threading Thus far, our web server has been entirely reactive, which can be a bottleneck for scalable systems Multi-threading is essential to develop scalable & robust networked applications, particularly servers The next group of slides present a domain analysis of concurrency design dimensions that address the policies & mechanisms governing the proper use of processes, threads, & synchronizers We outline the following design dimensions in this discussion: Iterative versus concurrent versus reactive servers Processes versus threads Process/thread spawning strategies User versus kernel versus hybrid threading models Time-shared versus real-time scheduling classes

Iterative vs. Concurrent Servers: 

Iterative vs. Concurrent Servers Iterative/reactive servers handle each client request in its entirety before servicing subsequent requests Best suited for short-duration or infrequent services Concurrent servers handle multiple requests from clients simultaneously Best suited for I/O-bound services or long-duration services Also good for busy servers

Multiprocessing vs. Multithreading: 

Multiprocessing vs. Multithreading A process provides the context for executing program instructions Each process manages certain resources (such as virtual memory, I/O handles, & signal handlers) & is protected from other OS processes via an MMU IPC between processes can be complicated & inefficient A thread is a sequence of instructions in the context of a process Each thread manages certain resources (such as runtime stack, registers, signal masks, priorities, & thread-specific data) Threads are not protected from other threads IPC between threads can be more efficient than IPC between processes

Thread Pool Eager Spawning Strategies: 

Thread Pool Eager Spawning Strategies This strategy prespawns one or more OS processes or threads at server creation time These``warm-started'' execution resources form a pool that improves response time by incurring service startup overhead before requests are serviced Two general types of eager spawning strategies are shown below: These strategies based on Half-Sync/Half-Async & Leader/Followers patterns

Thread-per-Request On-demand Spawning Strategy: 

Thread-per-Request On-demand Spawning Strategy On-demand spawning creates a new process or thread in response to the arrival of client connection and/or data requests Typically used to implement the thread-per-request & thread-per-connection models The primary benefit of on-demand spawning strategies is their reduced consumption of resources The drawbacks, however, are that these strategies can degrade performance in heavily loaded servers & determinism in real-time systems due to costs of spawning processes/threads & starting services

The N:1 & 1:1 Threading Models: 

The N:1 & 1:1 Threading Models OS scheduling ensures applications use host CPU resources suitably Modern OS platforms provide various models for scheduling threads A key difference between the models is the contention scope in which threads compete for system resources, particularly CPU time The two different contention scopes are shown below: Process contention scope (aka “user threading”) where threads in the same process compete with each other (but not directly with threads in other processes) System contention scope (aka “kernel threading”) where threads compete directly with other system-scope threads, regardless of what process they’re in

The N:M Threading Model: 

The N:M Threading Model Some operating systems (such as Solaris) offer a combination of the N:1 & 1:1 models, referred to as the ``N:M'‘ hybrid-threading model When an application spawns a thread, it can indicate in which contention scope the thread should operate The OS threading library creates a user-space thread, but only creates a kernel thread if needed or if the application explicitly requests the system contention scope When the OS kernel blocks an LWP, all user threads scheduled onto it by the threads library also block However, threads scheduled onto other LWPs in the process can continue to make progress

Scaling Up Performance via Threading : 

Scaling Up Performance via Threading Context HTTP runs over TCP, which uses flow control to ensure that senders do not produce data more rapidly than slow receivers or congested networks can buffer & process Since achieving efficient end-to-end quality of service (QoS) is important to handle heavy Web traffic loads, a Web server must scale up efficiently as its number of clients increases Problem Similarly, to improve QoS for all its connected clients, an entire Web server process must not block while waiting for connection flow control to abate so it can finish sending a file to a client Processing all HTTP GET requests reactively within a single-threaded process does not scale up, because each server CPU time-slice spends much of its time blocked waiting for I/O operations to complete

The Half-Sync/Half-Async Pattern: 

The Half-Sync/Half-Async Pattern The Half-Sync/Half-Async architectural pattern decouples async & sync service processing in concurrent systems, to simplify programming without unduly reducing performance This solution yields two benefits: Threads can be mapped to separate CPUs to scale up server performance via multi-processing Each thread blocks independently, which prevents a flow-controlled connection from degrading the QoS that other clients receive

Slide146: 

This pattern defines two service processing layers—one async & one sync—along with a queueing layer that allows services to exchange messages between the two layers Half-Sync/Half-Async Pattern Dynamics The pattern allows sync services, such as HTTP protocol processing, to run concurrently, relative both to each other & to async services, such as event demultiplexing

Applying Half-Sync/Half-Async Pattern in JAWS: 

Applying Half-Sync/Half-Async Pattern in JAWS <<get>> <<get>> <<get>> <<put>> <<ready to read>> Synchronous Service Layer Asynchronous Service Layer Queueing Layer Worker Thread 1 Worker Thread 3 ACE_Reactor Socket Event Sources Request Queue HTTP Acceptor HTTP Handlers, Worker Thread 2

Pros & Cons of Half-Sync/Half-Async Pattern: 

Pros & Cons of Half-Sync/Half-Async Pattern This pattern has three benefits: Simplification & performance The programming of higher-level synchronous processing services are simplified without degrading the performance of lower-level system services Separation of concerns Synchronization policies in each layer are decoupled so that each layer need not use the same concurrency control strategies Centralization of inter-layer communication Inter-layer communication is centralized at a single access point, because all interaction is mediated by the queueing layer This pattern also incurs liabilities: A boundary-crossing penalty may be incurred This overhead arises from context switching, synchronization, & data copying overhead when data is transferred between the sync & async service layers via the queueing layer Higher-level application services may not benefit from the efficiency of async I/O Depending on the design of operating system or application framework interfaces, it may not be possible for higher-level services to use low-level async I/O devices effectively Complexity of debugging & testing Applications written with this pattern can be hard to debug due its concurrent execution

Implementing a Synchronized Request Queue: 

Context The Half-Sync/Half-Async pattern contains a queue The JAWS Reactor thread is a ‘producer’ that inserts HTTP GET requests into the queue Worker pool threads are ‘consumers’ that remove & process queued requests <<get>> <<get>> <<get>> <<put>> Worker Thread 1 Worker Thread 3 ACE_Reactor Request Queue HTTP Acceptor HTTP Handlers, Worker Thread 2 Implementing a Synchronized Request Queue Problem A naive implementation of a request queue will incur race conditions or ‘busy waiting’ when multiple threads insert & remove requests e.g., multiple concurrent producer & consumer threads can corrupt the queue’s internal state if it is not synchronized properly Similarly, these threads will ‘busy wait’ when the queue is empty or full, which wastes CPU cycles unnecessarily

The Monitor Object Pattern: 

The Monitor Object Pattern This pattern synchronizes concurrent method execution to ensure that only one method at a time runs within an object It also allows an object’s methods to cooperatively schedule their execution sequences It’s instructive to compare Monitor Object pattern solutions with Active Object pattern solutions The key tradeoff is efficiency vs. flexibility

Monitor Object Pattern Dynamics: 

Monitor Object Pattern Dynamics the OS thread scheduler atomically reacquires the monitor lock the OS thread scheduler atomically releases the monitor lock Synchronized method invocation & serialization Synchronized method thread suspension Monitor condition notification Synchronized method thread resumption

Applying Monitor Object Pattern in JAWS: 

Applying Monitor Object Pattern in JAWS The JAWS synchronized request queue implements the queue’s not-empty & not-full monitor conditions via a pair of ACE wrapper facades for POSIX-style condition variables uses uses 2 Request Queue put() get() ACE_Thread_Mutex acquire() release() HTTP Handler ACE_Thread_Condition wait() signal() broadcast() Worker Thread <<put>> <<get>>

Pros & Cons of Monitor Object Pattern: 

Pros & Cons of Monitor Object Pattern This pattern provides two benefits: Simplification of concurrency control The Monitor Object pattern presents a concise programming model for sharing an object among cooperating threads where object synchronization corresponds to method invocations Simplification of scheduling method execution Synchronized methods use their monitor conditions to determine the circumstances under which they should suspend or resume their execution & that of collaborating monitor objects This pattern can also incur liabilities: The use of a single monitor lock can limit scalability due to increased contention when multiple threads serialize on a monitor object Complicated extensibility semantics These result from the coupling between a monitor object’s functionality & its synchronization mechanisms It is also hard to inherit from a monitor object transparently, due to the inheritance anomaly problem Nested monitor lockout This problem is similar to the preceding liability & can occur when a monitor object is nested within another monitor object

Minimizing Server Threading Overhead: 

Minimizing Server Threading Overhead When a connection request arrives, the operating system’s transport layer creates a new connected transport endpoint, encapsulates this new endpoint with a data-mode socket handle & passes the handle as the return value from accept() Context Socket implementations in certain multi-threaded operating systems provide a concurrent accept() optimization to accept client connection requests & improve the performance of Web servers that implement the HTTP 1.0 protocol as follows: The OS allows a pool of threads in a Web server to call accept() on the same passive-mode socket handle The OS then schedules one of the threads in the pool to receive this data-mode handle, which it uses to communicate with its connected client

Drawbacks with Half-Sync/Half-Async: 

Drawbacks with Half-Sync/Half-Async Problem Although Half-Sync/Half-Async threading model is more scalable than the purely reactive model, it is not necessarily the most efficient design CPU cache updates <<get>> <<get>> <<get>> <<put>> Worker Thread 1 Worker Thread 3 ACE_Reactor Request Queue HTTP Acceptor HTTP Handlers, Worker Thread 2 e.g., passing a request between the Reactor thread & a worker thread incurs: This overhead makes JAWS’ latency unnecessarily high, particularly on operating systems that support the concurrent accept() optimization

The Leader/Followers Pattern: 

The Leader/Followers Pattern The Leader/Followers architectural pattern (P2) provides an efficient concurrency model where multiple threads take turns sharing event sources to detect, demux, dispatch, & process service requests that occur on the event sources

Leader/Followers Pattern Dynamics: 

Leader/Followers Pattern Dynamics handle_events() new_leader() Leader thread demuxing Follower thread promotion Event handler demuxing & event processing Rejoining the thread pool promote_

Applying Leader/Followers Pattern in JAWS: 

Applying Leader/Followers Pattern in JAWS Two options: If platform supports accept() optimization then the Leader/Followers pattern can be implemented by the OS Otherwise, this pattern can be implemented as a reusable framework The Half-Sync/Half-Async design can reorder & prioritize client requests more flexibly, because it has a synchronized request queue implemented using the Monitor Object pattern It may be more scalable, because it queues requests in Web server virtual memory, rather than the OS kernel Although Leader/Followers thread pool design is highly efficient the Half-Sync/Half-Async design may be more appropriate for certain types of servers, e.g.:

Pros & Cons of Leader/Followers Pattern: 

Pros & Cons of Leader/Followers Pattern This pattern provides two benefits: Performance enhancements This can improve performance as follows: It enhances CPU cache affinity & eliminates the need for dynamic memory allocation & data buffer sharing between threads It minimizes locking overhead by not exchanging data between threads, thereby reducing thread synchronization It can minimize priority inversion because no extra queueing is introduced in the server It doesn’t require a context switch to handle each event, reducing dispatching latency Programming simplicity The Leader/Follower pattern simplifies the programming of concurrency models where multiple threads can receive requests, process responses, & demultiplex connections using a shared handle set This pattern also incur liabilities: Implementation complexity The advanced variants of the Leader/ Followers pattern are hard to implement Lack of flexibility In the Leader/ Followers model it is hard to discard or reorder events because there is no explicit queue Network I/O bottlenecks The Leader/Followers pattern serializes processing by allowing only a single thread at a time to wait on the handle set, which could become a bottleneck because only one thread at a time can demultiplex I/O events

Using Asynchronous I/O Effectively: 

Using Asynchronous I/O Effectively Context Synchronous multi-threading may not be the most scalable way to implement a Web server on OS platforms that support async I/O more efficiently than synchronous multi-threading AcceptEx() AcceptEx() AcceptEx() When these async operations complete, WinNT Delivers the associated completion events containing their results to the Web server Processes these events & performs the appropriate actions before returning to its event loop For example, highly-efficient Web servers can be implemented on Windows NT by invoking async Win32 operations that perform the following activities: Processing indication events, such as TCP CONNECT & HTTP GET requests, via AcceptEx() & ReadFile(), respectively Transmitting requested files to clients asynchronously via WriteFile() or TransmitFile()

The Proactor Pattern: 

The Proactor Pattern Problem Developing software that achieves the potential efficiency & scalability of async I/O is hard due to the separation in time & space of async operation invocations & their subsequent completion events

Proactor Pattern Dynamics: 

Proactor Pattern Dynamics Initiate operation Process operation Run event loop Generate & queue completion event Dequeue completion event & perform completion processing Note similarities & differences with the Reactor pattern, e.g.: Both process events via callbacks However, it’s generally easier to multi-thread a proactor

Applying the Proactor Pattern in JAWS: 

Applying the Proactor Pattern in JAWS The Proactor pattern structures the JAWS concurrent server to receive & process requests from multiple clients asynchronously JAWS HTTP components are split into two parts: Operations that execute asynchronously e.g., to accept connections & receive client HTTP GET requests The corresponding completion handlers that process the async operation results e.g., to transmit a file back to a client after an async connection operation completes

Proactive Connection Management & Data Transfer in JAWS: 

Proactive Connection Management & Data Transfer in JAWS

Pros & Cons of Proactor Pattern: 

Pros & Cons of Proactor Pattern This pattern offers five benefits: Separation of concerns Decouples application-independent async mechanisms from application-specific functionality Portability Improves application portability by allowing its interfaces to be reused independently of the OS event demuxing calls Decoupling of threading from concurrency The async operation processor executes long-duration operations on behalf of initiators so applications can spawn fewer threads Performance Avoids context switching costs by activating only those logical threads of control that have events to process Simplification of application synchronization If concrete completion handlers spawn no threads, application logic can be written with little or no concern for synchronization issues This pattern incurs some liabilities: Restricted applicability This pattern can be applied most efficiently if the OS supports asynchronous operations natively Complexity of programming, debugging, & testing It is hard to program applications & higher-level system services using asynchrony mechanisms, due to the separation in time & space between operation invocation & completion Scheduling, controlling, & canceling asynchronously running operations Initiators may be unable to control the scheduling order in which asynchronous operations are executed by an asynchronous operation processor

Efficiently Demuxing Asynchronous Operations & Completions: 

Efficiently Demuxing Asynchronous Operations & Completions Context In a proactive Web server async I/O operations will yield I/O completion event responses that must be processed efficiently Problem As little overhead as possible should be incurred to determine how the completion handler will demux & process completion events after async operations finish executing When a response arrives, the application should spend as little time as possible demultiplexing the completion event to the handler that will process the async operation’s response Together with each async operation that a client initiator invokes on a service, transmit information that identifies how the initiator should process the service’s response Return this information to the initiator when the operation finishes, so that it can be used to demux the response efficiently, allowing the initiator to process it accordingly

Asynchronous Completion Token Pattern: 

Asynchronous Completion Token Pattern

Applying the Asynchronous Completion Token Pattern in JAWS: 

Applying the Asynchronous Completion Token Pattern in JAWS

Pros & Cons of Asynchronous Completion Token Pattern: 

Pros & Cons of Asynchronous Completion Token Pattern This pattern has some liabilities: Memory leaks Memory leaks can result if initiators use ACTs as pointers to dynamically allocated memory & services fail to return the ACTs, for example if the service crashes Authentication When an ACT is returned to an initiator on completion of an asynchronous event, the initiator may need to authenticate the ACT before using it Application re-mapping If ACTs are used as direct pointers to memory, errors can occur if part of the application is re-mapped in virtual memory This pattern has four benefits: Simplified initiator data structures Initiators need not maintain complex data structures to associate service responses with completion handlers Efficient state acquisition ACTs are time efficient because they need not require complex parsing of data returned with the service response Space efficiency ACTs can consume minimal data space yet can still provide applications with sufficient information to associate large amounts of state to process asynchronous operation completion actions Flexibility User-defined ACTs are not forced to inherit from an interface to use the service’s ACTs

Enhancing Server (Re)Configurability (1/2): 

Enhancing Server (Re)Configurability (1/2) Certain factors are static, such as the number of available CPUs & operating system support for asynchronous I/O Other factors are dynamic, such as system workload Context The implementation of certain web server components depends on a variety of factors: Problem Prematurely committing to a particular web server component configuration is inflexible & inefficient: No single web server configuration is optimal for all use cases Certain design decisions cannot be made efficiently until run-time

Enhancing Server (Re)Configurability (2/2): 

Enhancing Server (Re)Configurability (2/2) This pattern allows an application to link & unlink its component implementations at run-time Thus, new & enhanced services can be added without having to modify, recompile, statically relink, or shut down & restart a running application

Component Configurator Pattern Dynamics: 

Component Configurator Pattern Dynamics run_component() run_component() fini() remove() remove() fini() Comp. A Concrete Comp. B Concrete Comp. A Concrete Comp. B Component initialization & dynamic linking Component processing Component termination & dynamic unlinking

Applying the Component Configurator Pattern to Content Servers: 

Applying the Component Configurator Pattern to Content Servers <<contains>> components * Component Configurator Component Repository LRU File Cache LFU File Cache Component init() fini() suspend() resume() info() For example, a content server can apply the Component Configurator pattern to configure various Cached Virtual Filesystem strategies e.g., least-recently used (LRU) or least-frequently used (LFU) Image servers can use the Component Configurator pattern to dynamically optimize, control, & reconfigure the behavior of its components at installation-time or during run-time

Reconfiguring JAWS: 

Reconfiguring JAWS Image servers can also be reconfigured dynamically to support new components & new component implementations

Pros & Cons of Component Configurator Pattern: 

Pros & Cons of Component Configurator Pattern This pattern offers four benefits: Uniformity By imposing a uniform configuration & control interface to manage components Centralized administration By grouping one or more components into a single administrative unit that simplifies development by centralizing common component initialization & termination activities Modularity, testability, & reusability Application modularity & reusability is improved by decoupling component implementations from the manner in which the components are configured into processes Configuration dynamism & control By enabling a component to be dynamically reconfigured without modifying, recompiling, statically relinking existing code & without restarting the component or other active components with which it is collocated This pattern also incurs liabilities: Lack of determinism & ordering dependencies This pattern makes it hard to determine or analyze the behavior of an application until its components are configured at run-time Reduced security or reliability An application that uses the Component Configurator pattern may be less secure or reliable than an equivalent statically-configured application Increased run-time overhead & infrastructure complexity By adding levels of abstraction & indirection when executing components Overly narrow common interfaces The initialization or termination of a component may be too complicated or too tightly coupled with its context to be performed in a uniform manner

Transparently Parameterizing Synchronization into Components: 

Transparently Parameterizing Synchronization into Components Context The various concurrency patterns described earlier impact component synchronization strategies in various ways e.g.,ranging from no locks to readers/writer locks In general, components must run efficiently in a variety of concurrency models Problem It should be possible to customize JAWS component synchronization mechanisms according to the requirements of particular application use cases & configurations Hard-coding synchronization strategies into component implementations is inflexible Maintaining multiple versions of components manually is not scalable Each type objectifies a particular synchronization strategy, such as a mutex, readers/writer lock, semaphore, or ‘null’ lock Instances of these pluggable types can be defined as objects contained within a component, which then uses these objects to synchronize its method implementations efficiently

Applying Polymorphic Strategized Locking in JAWS: 

Applying Polymorphic Strategized Locking in JAWS class File_Cache { public: // Constructor. File_Cache (Lock &l): lock_ (&l) { } // A method. const void *lookup (const string &path) const { lock_->acquire_read (); // Implement the <lookup> method. lock_->release (); } // ... private: // The polymorphic strategized locking object. mutable Lock *lock_; // Other data members & methods go here... }; Polymorphic Strategized Locking class Lock { public: // Acquire & release the lock. virtual void acquire () = 0; virtual void acquire_read () = 0; virtual void release () = 0; // ... }; class Thread_Mutex : public Lock { // ... };

Applying Parameterized Strategized Locking in JAWS: 

Applying Parameterized Strategized Locking in JAWS template <class LOCK> class File_Cache { public: // A method. const void *lookup (const string &path) const { lock_.acquire_read (); // Implement the <lookup> method. lock_.release (); } // ... private: // The polymorphic strategized locking object. mutable LOCK lock_; // Other data members & methods go here... }; Parameterized Strategized Locking Single-threaded file cache. typedef File_Cache<ACE_Null_Mutex> Content_Cache; Multi-threaded file cache using a thread mutex. typedef File_Cache<ACE_Thread_Mutex> Content_Cache; Multi-threaded file cache using a readers/writer lock. typedef File_Cache<ACE_RW_Mutex> Content_Cache; Note that the various locks need not inherit from a common base class or use virtual methods!

Pros & Cons of the Strategized Locking Pattern: 

Pros & Cons of the Strategized Locking Pattern This pattern provides three benefits: Enhanced flexibility & customization It is straightforward to configure & customize a component for certain concurrency models because the synchronization aspects of components are strategized Decreased maintenance effort for components It is straightforward to add enhancements & bug fixes to a component because there is only one implementation, rather than a separate implementation for each concurrency model Improved reuse Components implemented using this pattern are more reusable, because their locking strategies can be configured orthogonally to their behavior This pattern also incurs liabilities: Obtrusive locking If templates are used to parameterize locking aspects this will expose the locking strategies to application code Over-engineering Externalizing a locking mechanism by placing it in a component’s interface may actually provide too much flexibility in certain situations e.g., inexperienced developers may try to parameterize a component with the wrong type of lock, resulting in improper compile- or run-time behavior

Ensuring Locks are Released Properly: 

Ensuring Locks are Released Properly Context Concurrent applications, such as JAWS, contain shared resources that are manipulated by multiple threads concurrently

Applying the Scoped Locking Idiom in JAWS: 

Applying the Scoped Locking Idiom in JAWS template <class LOCK> class File_Cache { public: // A method. const void *lookup (const string &path) const { // Use Scoped Locking idiom to acquire // & release the <lock_> automatically. ACE_Read_Guard<LOCK> guard (*lock); // Implement the <lookup> method. // lock_ released automatically… } Applying the ACE_Guard

Pros & Cons of the Scoped Locking Idiom: 

Pros & Cons of the Scoped Locking Idiom This idiom has one benefit: Increased robustness This idiom increases the robustness of concurrent applications by eliminating common programming errors related to synchronization & multi-threading By applying the Scoped Locking idiom, locks are acquired & released automatically when control enters & leaves critical sections defined by C++ method & block scopes This idiom also has liabilities: Potential for deadlock when used recursively If a method that uses the Scoped Locking idiom calls itself recursively, ‘self-deadlock’ will occur if the lock is not a ‘recursive’ mutex Limitations with language-specific semantics The Scoped Locking idiom is based on a C++ language feature & therefore will not be integrated with operating system-specific system calls Thus, locks may not be released automatically when threads or processes abort or exit inside a guarded critical section Likewise, they will not be released properly if the standard C longjmp() function is called because this function does not call the destructors of C++ objects as the run-time stack unwinds

Minimizing Unnecessary Locking (1/2): 

Minimizing Unnecessary Locking (1/2) Context Components in multi-threaded applications that contain intra-component method calls Components that have applied the Strategized Locking pattern Problem Thread-safe components should be designed to avoid unnecessary locking Thread-safe components should be designed to avoid “self-deadlock” template <class LOCK> class File_Cache { public: const void *lookup (const string &path) const { ACE_Read_Guard<LOCK> guard (lock_); const void *file_pointer = check_cache (path); if (file_pointer == 0) { insert (path); file_pointer = check_cache (path); } return file_pointer; } void insert (const string &path) { ACE_Read_Guard<LOCK> guard (lock_); // ... insert <path> into cache... } private: mutable LOCK lock_; const void *check_cache (const string &) const; }; Since File_Cache is a template we don’t know the type of lock used to parameterize it.

Minimizing Unnecessary Locking (2/2): 

Minimizing Unnecessary Locking (2/2) Interface methods check All interface methods, such as C++ public methods, should only acquire/release component lock(s), thereby performing synchronization checks at the ‘border’ of the component. Call implementation methods to do work Don’t call interface This pattern structures all components that process intra-component method invocations according two design conventions: Implementation methods trust Implementation methods, such as C++ private & protected methods, should only perform work when called by interface methods. Only call other implementation methods

Applying the Thread-safe Interface Pattern in JAWS: 

Applying the Thread-safe Interface Pattern in JAWS template <class LOCK> class File_Cache { public: // Return a pointer to the memory-mapped file associated with // <path> name, adding it to the cache if it doesn’t exist. const void *lookup (const string &path) const { // Use Scoped Locking to acquire/release lock automatically. ACE_Read_Guard<LOCK> guard (lock_); return lookup_i (path); } private: mutable LOCK lock_; // The strategized locking object. // This implementation method doesn’t acquire or release // <lock_> & does its work without calling interface methods. const void *lookup_i (const string &path) const { const void *file_pointer = check_cache_i (path); if (file_pointer == 0) { // If <path> isn’t in cache, insert it & look it up again. insert_i (path); file_pointer = check_cache_i (path); // The calls to implementation methods <insert_i> & // <check_cache_i> assume that the lock is held & do work. } return file_pointer; Note fewer constraints on the type of LOCK…

Pros & Cons of the Thread-safe Interface Pattern: 

Pros & Cons of the Thread-safe Interface Pattern This pattern has some liabilities: Additional indirection & extra methods Each interface method requires at least one implementation method, which increases the footprint of the component & may also add an extra level of method-call indirection for each invocation Potential for misuse OO languages, such as C++ & Java, support class-level rather than object-level access control As a result, an object can bypass the public interface to call a private method on another object of the same class, thus bypassing that object’s lock Potential overhead This pattern prevents multiple components from sharing the same lock & prevents locking at a finer granularity than the component, which can increase lock contention This pattern has three benefits: Increased robustness This pattern ensures that self-deadlock does not occur due to intra-component method calls Enhanced performance This pattern ensures that locks are not acquired or released unnecessarily Simplification of software Separating the locking & functionality concerns can help to simplify both aspects

Synchronizing Singletons Correctly: 

Synchronizing Singletons Correctly Context JAWS uses various singletons to implement components where only one instance is required e.g., the ACE Reactor, the request queue, etc. Problem Singletons can be problematic in multi-threaded programs

The Double-checked Locking Optimization Pattern: 

The Double-checked Locking Optimization Pattern // Perform first-check to // evaluate ‘hint’. if (first_time_in is TRUE) { acquire the mutex // Perform double-check to // avoid race condition. if (first_time_in is TRUE) { execute the critical section set first_time_in to FALSE } release the mutex } class Singleton { public: static Singleton *instance () { // First check if (instance_ == 0) { Guard<Thread_Mutex> g(lock_); // Double check. if (instance_ == 0) instance_ = new Singleton; } return instance_; } private: static Singleton *instance_; static Thread_Mutex lock_; };

Applying the Double-Checked Locking Optimization Pattern in ACE: 

Applying the Double-Checked Locking Optimization Pattern in ACE template <class TYPE> class ACE_Singleton { public: static TYPE *instance () { // First check if (instance_ == 0) { // Scoped Locking acquires & release lock. ACE_Guard<ACE_Thread_Mutex> guard (lock_); // Double check instance_. if (instance_ == 0) instance_ = new TYPE; } return instance_; } private: static TYPE *instance_; static ACE_Thread_Mutex lock_; }; ACE defines a “singleton adapter” template to automate the double-checked locking optimization

Pros & Cons of the Double-Checked Locking Optimization Pattern: 

Pros & Cons of the Double-Checked Locking Optimization Pattern This pattern has two benefits: Minimized locking overhead By performing two first-time-in flag checks, this pattern minimizes overhead for the common case After the flag is set the first check ensures that subsequent accesses require no further locking Prevents race conditions The second check of the first-time-in flag ensures that the critical section is executed just once This pattern has some liabilities: Non-atomic pointer or integral assignment semantics If an instance_ pointer is used as the flag in a singleton implementation, all bits of the singleton instance_ pointer must be read & written atomically in a single operation If the write to memory after the call to new is not atomic, other threads may try to read an invalid pointer Multi-processor cache coherency Certain multi-processor platforms, such as the COMPAQ Alpha & Intel Itanium, perform aggressive memory caching optimizations in which read & write operations can execute ‘out of order’ across multiple CPU caches, such that the CPU cache lines will not be flushed properly if shared data is accessed without locks held

Logging Access Statistics Efficiently: 

Logging Access Statistics Efficiently Context Web servers often need to log certain information e.g., count number of times web pages are accessed Problem Having a central logging object in a multi-threaded server process can become a bottleneck e.g., due to synchronization required to serialize access by multiple threads

Logging Access Statistics Efficiently: 

Logging Access Statistics Efficiently Solution Apply the Thread-Specific Storage design pattern (P2) to allow multiple threads to use one ‘logically global’ access point to retrieve an object that is local to a thread, without incurring locking overhead on each object access errno is a good example of thread-specific storage

Thread-Specific Storage Pattern Dynamics: 

Thread-Specific Storage Pattern Dynamics

Applying the Thread-Specific Storage Pattern to JAWS: 

class Error_Logger { public: int last_error (); void log (const char *format, ...); }; Applying the Thread-Specific Storage Pattern to JAWS template <class TYPE> Class ACE_TSS { public: TYPE *operator->() const { TYPE *tss_data = 0; if (!once_) { ACE_Guard<ACE_Thread_Mutex> g (keylock_); if (!once_) { ACE_OS::thr_keycreate (&key_, &cleanup_hook); once_ = true; } } ACE_OS::thr_getspecific (key_, (void **) &tss_data); if (tss_data == 0) { tss_data = new TYPE; ACE_OS::thr_setspecific (key_, (void *) tss_data); } return tss_data; } private: mutable pthread_key_t key_; mutable bool once_; mutable ACE_Thread_Mutex keylock_; static void cleanup_hook (void *ptr); }; ACE_TSS <Error_Logger> my_logger; // ... if (recv (……) == -1 && my_logger->last_error () != EWOULDBLOCK) my_logger->log (“recv failed, errno = %d”, my_logger->last_error ()); };

Pros & Cons of the Thread-Specific Storage Pattern: 

Pros & Cons of the Thread-Specific Storage Pattern This pattern has four benefits: Efficiency It’s possible to implement this pattern so that no locking is needed to access thread-specific data Ease of use When encapsulated with wrapper facades & proxies, thread-specific storage is easy for application developers to use Reusability By combining this pattern with the Wrapper Façade pattern it’s possible to shield developers from non-portable OS platform characteristics Portability It’s possible to implement portable thread-specific storage mechanisms on most multi-threaded operating systems This pattern also has liabilities: It encourages use of thread-specific global objects Many applications do not require multiple threads to access thread-specific data via a common access point In this case, data should be stored so that only the thread owning the data can access it It obscures the structure of the system The use of thread-specific storage potentially makes an application harder to understand, by obscuring the relationships between its components It restricts implementation options Not all languages support parameterized types or smart pointers, which are useful for simplifying the access to thread-specific data

Additional Information: 

Patterns & frameworks for concurrent & networked objects www.cs.wustl.edu/~schmidt/POSA/ ACE & TAO open-source middleware www.cs.wustl.edu/~schmidt/ACE.html www.cs.wustl.edu/~schmidt/TAO.html ACE research papers www.cs.wustl.edu/~schmidt/ACE-papers.html Extended ACE & TAO tutorials UCLA extension, Feb, 2007 www.cs.wustl.edu/~schmidt/UCLA.html ACE books www.cs.wustl.edu/~schmidt/ACE/ Additional Information

Example: Applying Patterns to Real-time CORBA: 

Example: Applying Patterns to Real-time CORBA Patterns are used throughout The ACE ORB (TAO) Real-time CORBA implementation to codify expert knowledge & to generate the ORB’s software architecture by capturing recurring structures & dynamics & resolving common design forces www.cs.wustl.edu/~schmidt/POSA

R&D Context for ACE+TAO+CIAO: 

R&D Context for ACE+TAO+CIAO Our R&D focus: Advancing distruptive technologies to commoditize distributed real-time & embedded (DRE) systems

TAO–The ACE ORB: 

TAO–The ACE ORB More than 500 Ksloc (C++) Open-source Based on ACE wrapper facades & frameworks Available on Unix, Win32, MVS, QNX, VxWorks, LynxOS, VMS, etc. Thousands of users around the world Objective: Advance technology to simplify development of distributed, real-time, & embedded (DRE) systems Approach: Use standard OO techologies & patterns Commercially supported by OCI (www.theaceorb.com) PrismTech (www.prismtechnologies.com) Remedy (www.remedy.nl) etc.

The Evolution of TAO: 

The Evolution of TAO TAO ORB Largely compliant with CORBA 3.0 No DCOM bridge ;-) Pattern-oriented software architecture www.posa.uci.edu Key capabilities QoS-enabled Highly configurable Pluggable protocols IIOP/UIOP DIOP Shared memory SSL MIOP SCIOP TAO can be downloaded from deuce.doc.wustl.edu/Download.html

The Evolution of TAO: 

The Evolution of TAO RT-CORBA Portable priorities Protocol properties Standard synchronizers Explicit binding mechanisms Thread pools TAO 1.5 (Mar ’06) Current “official” release of TAO Heavily tested & optimized Baseline for next OCI & PrismTech supported releases www.dre.vanderbilt.edu/ scoreboard ZEN RT-CORBA/RT-Java Alpha now available www.zen.uci.edu RT-CORBA 1.0

The Evolution of TAO: 

The Evolution of TAO A/V STREAMING DYNAMIC/STATIC SCHEDULING A/V Streaming Service QoS mapping QoS monitoring QoS adaptation ACE QoS API (AQoSA) GQoS/RAPI & DiffServ IntServ integrated with A/V Streaming & QuO DiffServ integrated with ORB RT-CORBA 1.0 Static Scheduling (1.0) Rate monotonic analysis Dynamic Scheduling (1.2) Earliest deadline first Minimum laxity first Maximal urgency first Hybrid Dynamic/Static Demo in WSOA Kokyu integrated in Summer 2003

The Evolution of TAO: 

The Evolution of TAO DYNAMIC/STATIC SCHEDULING FT-CORBA & LOAD BALANCING FT-CORBA (DOORS) Entity redundancy Multiple models Cold passive Warm passive IOGR HA/FT integrated by Winter 2004 Load Balancing Static & dynamic Integrated in TAO 1.3 De-centralized LB OMG LB specification SSL Support Integrity Confidentiality Authentication (limited) Security Service (CSIv2) Authentication Access control Non-repudiation Audit Beta by Winter 2004 A/V STREAMING SECURITY RT-CORBA 1.0

The Evolution of TAO: 

The Evolution of TAO NOTIFICATIONS A/V STREAMING SECURITY TRANSACTIONS DYNAMIC/STATIC SCHEDULING FT-CORBA & LOAD BALANCING Notification Service Structured events Event filtering QoS properties Priority Expiry times Order policy Compatible w/Events Real-time Notification Service Summer 2003 Object Transaction Service Encapsulates RDBMs www.xots.org RT-CORBA 1.0

The Evolution of TAO: 

The Evolution of TAO NOTIFICATIONS A/V STREAMING SECURITY TRANSACTIONS DYNAMIC/STATIC SCHEDULING FT-CORBA & LOAD BALANCING CORBA Component Model (CIAO) Extension Interfaces Component navigation Standardized life-cycles QoS-enabled containers Reflective collocation Implements the OMG Deployment & Configuration specification First major release (1.0) by Winter 2005 RT-CORBA 1.0

The Road Ahead (1/3): 

Middleware Middleware Services DRE Applications Operating Sys & Protocols Hardware & Networks Limit to how much application functionality can be factored into reusable COTS middleware, which impedes product-line architectures Middleware itself has become extremely complicated to use & provision statically & dynamically Component-based DRE systems are very complicated to deploy & configure There are now multiple middleware technologies to choose from The Road Ahead (1/3)

The Road Ahead (2/3): 

Develop, validate, & standardize model-driven development (MDD) software technologies that: Model Analyze Synthesize & Provision multiple layers of middleware & application components that require simultaneous control of multiple quality of service properties end-to-end Partial specialization is essential for inter-/intra-layer optimization & advanced product-line architectures <CONFIGURATION_PASS> <HOME> <…> <COMPONENT> <ID> <…></ID> <EVENT_SUPPLIER> <…events this component supplies…> </EVENT_SUPPLIER> </COMPONENT> </HOME> </CONFIGURATION_PASS> Goal is not to replace programmers per se – it is to provide higher-level domain-specific languages for middleware/application developers & users The Road Ahead (2/3)

The Road Ahead (3/3): 

Our MDD toolsuite is called CoSMIC (“Component Synthesis using Model Integrated Computing”) www.dre.vanderbilt.edu/cosmic The Road Ahead (3/3)

Concluding Remarks: 

Concluding Remarks Researchers & developers of distributed applications face common challenges R&D Synergies Patterns, frameworks, & components help to resolve these challenges These techniques can yield efficient, scalable, predictable, & flexible middleware & applications e.g., connection management, service initialization, error handling, flow & congestion control, event demuxing, distribution, concurrency control, fault tolerance synchronization, scheduling, & persistence