NESCDPA May31 07

Linking Programming models between Grids, Web 2.0 and Multicore : 

1 Linking Programming models between Grids, Web 2.0 and Multicore Distributed Programming Abstractions Workshop NESC Edinburgh UK May 31 2007 Geoffrey Fox Computer Science, Informatics, Physics Pervasive Technology Laboratories Indiana University Bloomington IN 47401 gcf@indiana.edu http://www.infomall.org

Points in Talk I: 

2 Points in Talk I All parallel programming projects more or less fail All distributed programming projects report success There are several hundred in Grid workflow area alone Few constraints on distributed programming Composition (in distributed computing) v decomposition (in parallel computing) There is not much difference between distributed programming and a key paradigm of parallel computing (functional parallelism) Pervasive use of  64 core chips in the future will often require one to build a Grid on a chip i.e. to execute a traditional distributed application on a chip XML is a pretty dubious syntax for expressing programs Web 2.0 is pretty scruffy but there are some large companies and many users behind it. Web 2.0 and Grids will converge and features of both will survive or disappear in merged environment Web 2.0 has a more plausible approach to distributed programming than Web Services/Grids Dominant Distributed Programming models will support Multicore, Web 2.0 and Grids

Some More points: 

Some More points Services could be universal abstraction in parallel and distributed computing Whereas objects could not be universal so perhaps should move away from their use Gateways/Portals (Portlets, Widgets, Gadgets) are natural user (application usage) interface to a collection of services Important Data (SQL, WFS, RSS Feeds) abstractions Divide Parallel Programming Run-time (matching application structure) into 3 or 4 Broad classes Inter-entity communication time characteristic of different programming model 1-5 µs for MPI/Thread switching to 1-1000 milliseconds for services on the Grid and 25 µs for services inside a chip Multicore Commodity Programming Model Marine corps write libraries in 'HLA++', MPI or dynamic threads (internally one microsecond latency) expressed as services Services composed/mashuped by 'millions' Many composition (coordination) or mashup approaches Functional (cf. Google Map Reduce for data transformations) Dataflow Workflow Visual Script The difficulties of making effective use of multicore chips will so great that it will be main driver of new programming environments Microsoft CCR DSS is good example of unification of parallel and distributed computing

Some Details: 

Some Details See http://www.slideshare.net/Foxsden or more conventionally Web 2.0 and Grid Tutorial http://grids.ucs.indiana.edu/ptliupages/presentations/CTSpartIMay21-07.ppt http://grids.ucs.indiana.edu/ptliupages/presentations/Web20Tutorial_CTS.ppt Multicore and Parallel Computing Tutorial http://grids.ucs.indiana.edu/ptliupages/presentations/PC2007/index.html 'Web 2.0' citation site http://www.connotea.org/user/crmc

Web 2.0 and Web Services I: 

Web 2.0 and Web Services I Web Services have clearly defined protocols (SOAP) and a well defined mechanism (WSDL) to define service interfaces There is good .NET and Java support The so-called WS-* (WS-Nightmare) specifications provide a rich sophisticated but complicated standard set of capabilities for security, fault tolerance, meta-data, discovery, notification etc. 'Narrow Grids' build on Web Services and provide a robust managed environment with growing adoption in Enterprise systems and distributed science (e-Science) We can use the term Grids strictly as Narrow Grids that are collections of Web Services (or even more strictly OGSA Grids) or just call any collections of services as 'Broad Grids' which actually is quite often done Web 2.0 supports a similar architecture to Web services but has developed in a more chaotic but remarkably successful fashion with a service architecture with a variety of protocols including those of Web and Grid services Over 400 Interfaces defined at http://www.programmableweb.com/apis One can easily combine SOAP (Web Service) based services/systems with HTTP messages but the 'lowest common denominator' suggests additional structure/complexity of SOAP will not easily survive

Web 2.0 and Web Services II: 

Web 2.0 and Web Services II Web 2.0 also has many well known capabilities with Google Maps and Amazon Compute/Storage services of clear general relevance There are also Web 2.0 services supporting novel collaboration modes and user interaction with the web as seen in social networking sites and portals such as: MySpace, YouTube, Connotea, Slideshare …. I once thought Web Services were inevitable but this is no longer clear to me Web services are complicated, slow and non functional WS-Security is unnecessarily slow and pedantic (canonicalization of XML) WS-RM (Reliable Messaging) seems to have poor adoption and doesn’t work well in collaboration WSDM (distributed management) specifies a lot There are de facto standards like Google Maps and powerful suppliers like Google which 'define the rules'

Attack of the Killer Multicores: 

Attack of the Killer Multicores 7

Grids meet Multicore Systems: 

Grids meet Multicore Systems 8

Grid versus Multicore Applications: 

Grid versus Multicore Applications 9

Slide10: 

10 What is …? What if …? Is it …? Recognition Mining Synthesis Create a model instance RMS: Recognition Mining Synthesis Model-based multimodal recognition Find a model instance Model Real-time analytics on dynamic, unstructured, multimodal datasets Photo-realism and physics-based animation Model-less Real-time streaming and transactions on static – structured datasets Very limited realism Intel has probably most sophisticated analysis of future 'killer' multicore applications – they are 'just' standard Grid and parallel computing

Slide11: 

11 What is a tumor? Is there a tumor here? What if the tumor progresses? It is all about dealing efficiently with complex multimodal datasets Recognition Mining Synthesis Images courtesy: http://splweb.bwh.harvard.edu:8000/pages/images_movies.html

Intel’s Application Stack: 

PC07Intro gcf@indiana.edu 12 Intel’s Application Stack

Role of Data in Grid/Multicore I: 

Role of Data in Grid/Multicore I 13

Role of Data in Grid/Multicore: 

Role of Data in Grid/Multicore 14

Multicore Programming Paradigms: 

PC07Intro gcf@indiana.edu 15 Multicore Programming Paradigms At a very high level, there are three or four broad classes of parallelism Coarse grain functional parallelism typified by workflow and often used to build composite 'metaproblems' whose parts are also parallel 'Compute-File', Database/Sensor, Community, Service, Pleasing Parallel (Master-worker) are sub-classses Large Scale loosely synchronous data parallelism where dynamic irregular work has clear synchronization points as in most large scale scientific and engineering problems Fine grain (asynchronous) thread parallelism as used in search algorithms which are often data parallel (over choices) but don’t have universal synchronization points Discrete Event Simulations are either a fourth class or a variant of thread parallelism

Data Parallel Time Dependence: 

Data Parallel Time Dependence A simple form of data parallel applications are synchronous with all elements of the application space being evolved with essentially the same instructions Such applications are suitable for SIMD computers and run well on vector supercomputers (and GPUs but these are more general than just synchronous) However synchronous applications also run fine on MIMD machines SIMD CM-2 evolved to MIMD CM-5 with same data parallel language CMFortran The iterative solutions to Laplace’s equation are synchronous as are many full matrix algorithms Synchronization on MIMD machines is accomplished by messaging It is automatic on SIMD machines! Application Time Application Space Synchronous Identical evolution algorithms

Local Messaging for Synchronization: 

Local Messaging for Synchronization MPI_SENDRECV is typical primitive Processors do a send followed by a receive or a receive followed by a send In two stages (needed to avoid race conditions), one has a complete left shift Often follow by equivalent right shift, do get a complete exchange This logic guarantees correctly updated data is sent to processors that have their data at same simulation time ……… 8 Processors Application and Processor Time Application Space Communication Phase

Loosely Synchronous Applications: 

Loosely Synchronous Applications This is most common large scale science and engineering and one has the traditional data parallelism but now each data point has in general a different update Comes from heterogeneity in problems that would be synchronous if homogeneous Time steps typically uniform but sometimes need to support variable time steps across application space – however ensure small time steps are t = (t1-t0)/Integer so subspaces with finer time steps do synchronize with full domain The time synchronization via messaging is still valid However one no longer load balances (ensure each processor does equal work in each time step) by putting equal number of points in each processor Load balancing although NP complete is in practice surprisingly easy Distinct evolution algorithms for each data point in each processor

MPI Futures?: 

MPI Futures? MPI likely to become more important as multicore systems become more common Should use MPI when MPI needed and use other messaging for other cases (such as linking services) where different features/performance appropriate MPI has too many primitives which will handicap broad implementation/adoption Perhaps only have one collective primitive like CCR which allows general collective operations to be built by user

Fine Grain Dynamic Applications: 

Fine Grain Dynamic Applications Here there is no natural universal ‘time’ as there is in science algorithms where an iteration number or Mother Nature’s time gives global synchronization Loose (zero) coupling or special features of application needed for successful parallelization In computer chess, the minimax scores at parent nodes provide multiple dynamic synchronization points

Computer Chess: 

Computer Chess Thread level parallelism unlike position evaluation parallelism used in other systems Competed with poor reliability and results in 1987 and 1988 ACM Computer Chess Championships

Discrete Event Simulations: 

Discrete Event Simulations These are familiar in military and circuit (system) simulations when one uses macroscopic approximations Also probably paradigm of most multiplayer Internet games/worlds Note Nature is perhaps synchronous when viewed quantum mechanically in terms of uniform fundamental elements (quarks and gluons etc.) It is loosely synchronous when considered in terms of particles and mesh points It is asynchronous when viewed in terms of tanks, people, arrows etc. Circuit simulations can be done loosely synchronously but inefficient as many inactive elements

Programming Models : 

Programming Models The three major models are supported by HPCS languages which are very interesting but too monolithic So the Fine grain thread parallelism and Large Scale loosely synchronous data parallelism styles are distinctive to parallel computing while Coarse grain functional parallelism of multicore overlaps with workflows from Grids and Mashups from Web 2.0 Seems plausible that a more uniform approach evolve for coarse grain case although this is least constrained of programming styles as typically latency issues are not critical Multicore would have strongest performance constraints Web 2.0 and Multicore the most important usability constraints A possible model for broad use of multicores is that the difficult parallel algorithms are coded as libraries (Fine grain thread parallelism and Large Scale loosely synchronous data parallelism styles) while the general user uses composes with visual interfaces, scripting and systems like Google MapReduce

Google MapReduceSimplified Data Processing on Large Clusters: 

PC07Intro gcf@indiana.edu 24 Google MapReduce Simplified Data Processing on Large Clusters http://labs.google.com/papers/mapreduce.html This is a dataflow model between services where services can do useful document oriented data parallel applications including reductions The decomposition of services onto cluster engines is automated The large I/O requirements of datasets changes efficiency analysis in favor of dataflow Services (count words in example) can obviously be extended to general parallel applications There are many alternatives to language expressing either dataflow and/or parallel operations and indeed one should support multiple languages in spirit of services

Programming Models: 

Programming Models The services and objects in distributed computing are usually 'natural' (come from application) whereas parts connected by MPI (or created by parallelizing compiler) come from 'artificial' decompositions and not naturally considered services Services in multicore (parallel computing) are original modules before decomposition and its these modules that coarse grain functional parallelism addresses Most of 'difficult' issues in parallel computing concern treatment of decomposition

Parallel Software Paradigms: Top Level: 

Parallel Software Paradigms: Top Level In the conventional two-level Grid/Web Service programming model, one programs each individual service and then separately programs their interaction This is Grid-aware Services programming model SAGA supports Grid-aware programs? This is generalized to multicore with 'Marine Corps' programming services for 'difficult' cases Loosely Synchronous Fine Grain threading Discrete Event Simulation 'Average' Programmer produces mashups or workflows from these parallelized services

The Marine Corps Lack of Programming Paradigm Library Model: 

The Marine Corps Lack of Programming Paradigm Library Model One could assume that parallel computing is 'just too hard for real people' and assume that we use a Marine Corps of programmers to build as libraries excellent parallel implementations of 'all' core capabilities e.g. the primitives identified in the Intel application analysis e.g. the primitives supported in Google MapReduce, HPF, PeakStream, Microsoft Data Parallel .NET etc. These primitives are orchestrated (linked together) by overall frameworks such as workflow or mashups The Marine Corps probably is content with efficient rather than easy to use programming models

Component Parallel and Program Parallel: 

Component Parallel and Program Parallel Component parallel paradigm is where one explicitly programs the different parts of a parallel application with the linkage either specified externally as in workflow or in components themselves as in most other component parallel approaches In Grids, components are natural In Parallel computing, components are produced by decomposition In the program parallel paradigm, one writes a single program to describe the whole application and some combination of compiler and runtime breaks up the program into the multiple parts that execute in parallel Note that a program parallel approach will often call a built in runtime library written in component parallel fashion A parallelizing compiler could call an MPI library routine Could perhaps better call 'Program Parallel' as 'Implicitly Parallel' and 'Component Parallel' as 'Explicitly Parallel'

Component Parallel and Program Parallel: 

Component Parallel and Program Parallel Program Parallel approaches include Data structure parallel as in Google MapReduce, HPF (High Performance Fortran), HPCS (High-Productivity Computing Systems) or 'SIMD' co-processor languages (PeakStream, ClearSpeed and Microsoft Data Parallel .NET) Parallelizing compilers including OpenMP annotation Note OpenMP and HPF have failed in some sense for large scale parallel computing (writing algorithm in standard sequential languages throws away information needed for parallelization) Component Parallel approaches include MPI (and related systems like PVM) parallel message passing PGAS (Partitioned Global Address Space CAF, UPC, Titanium, HPJava ) C++ futures and active objects CSP … Microsoft CCR and DSS Workflow and Mashups Discrete Event Simulation

Why people like MPI!: 

Why people like MPI! Jason J Beech-Brandt, and Andrew A. Johnson, at AHPCRC Minneapolis BenchC is unstructured finite element CFD Solver Looked at OpenMP on shared memory Altix with some effort to optimize Optimized UPC on several machines MPI always good but other approaches erratic Other studies reach similar conclusions?

Web 2.0 Systems are Portals, Services, Resources: 

Web 2.0 Systems are Portals, Services, Resources Captures the incredible development of interactive Web sites enabling people to create and collaborate

Mashups v Workflow?: 

32 Mashups v Workflow? Mashup Tools are reviewed at http://blogs.zdnet.com/Hinchcliffe/?p=63 Workflow Tools are reviewed by Gannon and Fox http://grids.ucs.indiana.edu/ptliupages/publications/Workflow-overview.pdf Both include scripting in PHP, Python, sh etc. as both implement distributed programming at level of services Mashups use all types of service interfaces and do not have the potential robustness (security) of Grid service approach Typically 'pure' HTTP (REST)

Web 2.0 APIs: 

Web 2.0 APIs http://www.programmableweb.com/apis has (May 14 2007) 431 Web 2.0 APIs with GoogleMaps the most often used in Mashups This site acts as a 'UDDI' for Web 2.0

The List of Web 2.0 API’s: 

The List of Web 2.0 API’s Each site has API and its features Divided into broad categories Only a few used a lot (42 API’s used in more than 10 mashups) RSS feed of new APIs Amazon S3 growing in popularity

APIs/Mashups per Protocol Distribution: 

APIs/Mashups per Protocol Distribution Number of Mashups Number of APIs

4 more Mashups each day: 

4 more Mashups each day For a total of 1906 April 17 2007 (4.0 a day over last month) Note ClearForest runs Semantic Web Services Mashup competitions (not workflow competitions) Some Mashup types: aggregators, search aggregators, visualizers, mobile, maps, games

Implication for Grid Technology of Multicore and Web 2.0 I: 

Implication for Grid Technology of Multicore and Web 2.0 I 37

Implication for Grid Technology of Multicore and Web 2.0 II: 

Implication for Grid Technology of Multicore and Web 2.0 II 38

Implication for Grid Technology of Multicore and Web 2.0 III: 

Implication for Grid Technology of Multicore and Web 2.0 III 39

The Ten areas covered by the 60 core WS-* Specifications : 

The Ten areas covered by the 60 core WS-* Specifications

WS-* Areas and Web 2.0 : 

WS-* Areas and Web 2.0

WS-* Areas and Multicore : 

WS-* Areas and Multicore

CCR as an example of a Cross Paradigm Run Time: 

CCR as an example of a Cross Paradigm Run Time Naturally supports fine grain thread switching with message passing with around 4 microsecond latency for 4 threads switching to 4 others on an AMD PC with C#. Threads spawned – no rendezvous Has around 50 microsecond latency for coarse grain service interactions with DSS extension which supports Web 2.0 style messaging MPI Collectives – Shift and Exchange vary from 10 to 20 microsecond latency in rendezvous mode Not as good as best MPI’s but managed code and supports Grids Web 2.0 and Parallel Computing …… See http://grids.ucs.indiana.edu/ptliupages/publications/CCRApril16open.pdf

Microsoft CCR: 

PC07Intro gcf@indiana.edu 44 Microsoft CCR Supports exchange of messages between threads using named ports FromHandler: Spawn threads without reading ports Receive: Each handler reads one item from a single port MultipleItemReceive: Each handler reads a prescribed number of items of a given type from a given port. Note items in a port can be general structures but all must have same type. MultiplePortReceive: Each handler reads a one item of a given type from multiple ports. JoinedReceive: Each handler reads one item from each of two ports. The items can be of different type. Choice: Execute a choice of two or more port-handler pairings Interleave: Consists of a set of arbiters (port -- handler pairs) of 3 types that are Concurrent, Exclusive or Teardown (called at end for clean up). Concurrent arbiters are run concurrently but exclusive handlers are http://msdn.microsoft.com/robotics/

Slide45: 

Overhead (latency) of AMD 4-core PC with 4 execution threads on MPI style Rendezvous Messaging for Shift and Exchange implemented either as two shifts or as custom CCR pattern. Compute time is 10 seconds divided by number of stages Stages (millions) Time Microseconds

Slide46: 

Overhead (latency) of INTEL 8-core PC with 8 execution threads on MPI style Rendezvous Messaging for Shift and Exchange implemented either as two shifts or as custom CCR pattern. Compute time is 15 seconds divided by number of stages Stages (millions) Time Microseconds

Timing of HP Opteron Multicore as a function of number of simultaneous two-way service messages processed (November 2006 DSS Release): 

PC07Intro gcf@indiana.edu 47 Timing of HP Opteron Multicore as a function of number of simultaneous two-way service messages processed (November 2006 DSS Release) CGL Measurements of Axis 2 shows about 500 microseconds – DSS is 10 times better DSS Service Measurements