logging in or signing up ovms hist Goldye Download Post to : URL : Related Presentations : Share Add to Flag Embed Email Send to Blogs and Networks Add to Channel Uploaded from authorPOINTLite Insert YouTube videos in PowerPont slides with aS Desktop Copy embed code: (To copy code, click on the text box) Embed: URL: Thumbnail: WordPress Embed Customize Embed The presentation is successfully added In Your Favorites. Views: 141 Category: Entertainment License: All Rights Reserved Like it (0) Dislike it (0) Added: November 07, 2007 This Presentation is Public Favorites: 0 Presentation Description No description available. Comments Posting comment... Premium member Presentation Transcript OpenVMS Technology Review: OpenVMS Technology Review John Wisniewski OpenVMS Consultant wisniewski@mail.dec.com http://www.montagar.com/~johnw/ Slide2: www.digital.com www.montagar.com/dfwlug DECUS THE DFWLUG DALLAS/FT WORTHOpenVMS: OpenVMS When your system absolutely has to runWhat is OpenVMS?: What is OpenVMS? The dominant force in the minicomputer revolution Rated as the #1 OS in the healthcare industry (1996) The first OS to run 24x365 (1985) and still the leader The first OS to scale from desktop to data center (1987) The first disaster tolerant OS (1988) and still the leader The first OS to ship CORBA, one of the first to ship DCE, and the first non-UNIX OS to ship POSIX and SPEC 11/70 The first OS to run 3-tier client/server and still the leader The second OS to run 64-bits The most highly scalable OS in the world today OpenVMS History - 20th Anniversary : OpenVMS History - 20th Anniversary 1976 Programmable Data Processors 16-Bit PDP 1977 Virtual Address eXtened 32-Bit VAX 1977 Aug Virtual Memory System v1.0 shipped on on VAX 11/780 hw /VMS sw 1983 First ship VMS Clusters VMS 4.0OpenVMS History - 20th Anniversary : OpenVMS History - 20th Anniversary 1986 First ship of single chip 32-Bit single chip microprocessor MicroVAX II with VMS v4.2 1988 VMS supports first Symmetric Multiprocessing Processing software for general purpose computing on VAX 6000s VMS v5.0 OpenVMS History - 20th Anniversary : OpenVMS History - 20th Anniversary 1990 Display list DECwindows GUI (VWS) is replaced by DECwindows Motif with VMS v5.2 1991 Disaster Tolerant Clusters 150 miles apart 1992 VMS becomes OpenVMS with Posix and Spec 11/70 compliance X/open XPG3-XPG/4 Branded just like Unix 1992 OpenVMS Alpha 64bit support first delivered in Feb OpenVMS History - 20th Anniversary : OpenVMS History - 20th Anniversary 1993 OpenVMS v5.5-2 (Oldest Y2K certified OS) 1994 OpenVMS v6.1 VAX 32-bit and Alpha 64-bit functional equivilency, cross architecture clustering 1995 OpenVMS v6.2 SCSI clusters 1995 OpenVMS / WNT Affinity Program Announced in May 1996 OpenVMS v7.0 Alpha 64-bit differenciators from VAX rollout OpenVMS History - 20th Anniversary : OpenVMS History - 20th Anniversary 1997 OpenVMS v7.1 Alpha 64-bit applications Oracle 7.3 VLM/VLDB 1997 OpenVMS WNT Affinity program Announced Wave 4 with ISG Navigator and over thirty products for Windows Interoperablity and Client / Server Support 1997 Disaster Tolerant Cluster 540 Miles, Memory Channel (100Mbyte/Sec) Clusters 1997 OpenVMS / Galaxy Software demonstrated virtual machine, OpenVMS Clusters in a box.You’re going to have to answer some questions:: You’re going to have to answer some questions: Is OpenVMS a good long-term strategic investment? Is OpenVMS also strategic to Oracle?OpenVMS Business Summary: OpenVMS Business Summary FY98 - $70M budget 500 person Engineering organization Oracle 12 engineers onsite $1.5M funding #1 Operating System in Healthcare* 23% growth 15% new business * based on 1997 CHIME studyOpenVMS Engineering plan for the next five years: OpenVMS Engineering plan for the next five years Deliver innovative, leading edge Affinity systems in partnership with customers into the 21st century Be the ultimate high-end for Windows NT Be a superb Database and Internet server Raise the bar for availability Raise the bar for scalabilityOpenVMS and Windows NT Strategy: OpenVMS and Windows NT Strategy OpenVMS as Continuous Computing Server Enterprise strength Bulletproof 24x365 Highly scalable Proven in the most demanding environments 64 bits/VLM Highly Available Internet Servers Memory Channel Clusters Galaxy OpenVMS & Windows NT Connectivity Applications Development environment Momentum PC volume economics 3 Tier C/S ADE Seamless Access to Data Affinity System Management Windows Gateway to OVMS Distributed TP “The seamless integration of OpenVMS with Windows NT will enable OpenVMS customers to combine the proven enterprise strengths of OpenVMS with the new foundation of Windows NT applications.” - Bill Gates Microsoft/Digital, August 2, 1995…and a few more questions: …and a few more questions Does OpenVMS have the mission critical functionality we need? Does OpenVMS scale? Where is OpenVMS going in the future? What is “excellence” in availability in 1997?: What is “excellence” in availability in 1997? Paris: May 5, 1996OpenVMS clusters offer a long list of unmatched availability features : OpenVMS clusters offer a long list of unmatched availability features Up to 96 nodes and 28 petabytes in a cluster Distributed lock manager Cluster file system Multi-site deployment Wide-area shadowing Automatic failover Automated volume shadowing On-line back-up Rolling upgrades for: Hardware Systems software Application softwareOpenVMS certainly scales well enough as a stand-alone system: OpenVMS certainly scales well enough as a stand-alone system Rdb Gains Index scan - 55X Indexed Data Retrieval - 146X 3-way Join - 142X 5-way Join - 216X Single system record holder - 14,227 tpm-C - April 1997 64 Bits for VLM, VLDBAnd OpenVMS scales REALLY well in clusters…and can be multi-site disaster-tolerant besides: And OpenVMS scales REALLY well in clusters…and can be multi-site disaster-tolerant besides Internet NYC Paramus Single logical data imageLeadership functionality:Disaster-tolerant computing: Leadership functionality: Disaster-tolerant computing System Disk Data Disk Data Disk Shared Disks GIGAswitch System Disk Data Disk Data Disk Shared Disks GIGAswitch T3 or FDDI or ATM Shadowed Disks Site “A” Site “B” Easy to manage - Single system view Operation Management Software provides management of configuration from either site Automatic transfer in seconds in event of disaster Integrated, tested, installed system solution - 5 yrs of experience OpenVMS Galaxy Technology: OpenVMS Galaxy Technology Brings multi-system performance, availability & scalability to a single system An extension/evolution of current OpenVMS arch S/W arch that adapts to the way you do computing partitions a single machine into multiple SMP nodes environment is adaptive and dynamic multiple instances of OS execute in a single computer Addresses limitations in current system archs: clusters: lack of shared memory and interconnect SMP: S/W scalability and single system view MPP: limited applications, management Digital Proprietary and ConfidentialOpenVMS Galaxy Technology (cont): OpenVMS Galaxy Technology (cont) Consists of two components: APMP (Adaptive Partitioned Multi-Processing) model Galaxy Softwared Architecture to exploit APMP 4 Modes of Operation: Shared Nothing: partition CPU’s for dedicated apps Shared Partial: shared memory for VLDB Shared Everything: shared mem & cluster interconnect Adaptive Dynamic Resource Partitioning: migration resiliency Capabilities: 1-256 CPUs 300,000+ to 1,000,000 design goal dynamice resource allocations hot swap of CPU, mem, I/O Digital Proprietary and ConfidentialNeed bigger - Galaxy: Need bigger - Galaxy Node A I/O Node B I/O Node C I/O Shared Memory Digital ConfidentialGalaxy Software Architecture: Galaxy Software Architecture Node A I/O Node B I/O Node C I/O Shared Memory Digital Proprietary and ConfidentialGalaxy Software Architecture withAdaptive Dynamic Resource Partioning: Galaxy Software Architecture with Adaptive Dynamic Resource Partioning Node A I/O Node B I/O Node C I/O Shared Memory Digital Proprietary and ConfidentialGalaxy Implications: Galaxy Implications Shared memory Very large database cache Node reboot into “hot” cache DLM cache now local vs distributed Shared memory cluster interconnect SMP Improved SMP scaling Improved SMP affinity CPU Migration Partitioning of CPUs I/O Multiple primaries for INT Distribute INT to secondaries Can partition I/O workload Highly scalable I/O subsystem Reliability / Availability Hotswap (CPU, Memory, I/O) OS will fail out only the necessary subsystem Software failure will not compromise entire system OS Resilliency (dynamic reallocation or resources) Digital Proprietary and ConfidentialSlide27: www.digital.com www.openvms.digital.com www.montagar.com/dfwlug/ You do not have the permission to view this presentation. In order to view it, please contact the author of the presentation.
ovms hist Goldye Download Post to : URL : Related Presentations : Share Add to Flag Embed Email Send to Blogs and Networks Add to Channel Uploaded from authorPOINTLite Insert YouTube videos in PowerPont slides with aS Desktop Copy embed code: (To copy code, click on the text box) Embed: URL: Thumbnail: WordPress Embed Customize Embed The presentation is successfully added In Your Favorites. Views: 141 Category: Entertainment License: All Rights Reserved Like it (0) Dislike it (0) Added: November 07, 2007 This Presentation is Public Favorites: 0 Presentation Description No description available. Comments Posting comment... Premium member Presentation Transcript OpenVMS Technology Review: OpenVMS Technology Review John Wisniewski OpenVMS Consultant wisniewski@mail.dec.com http://www.montagar.com/~johnw/ Slide2: www.digital.com www.montagar.com/dfwlug DECUS THE DFWLUG DALLAS/FT WORTHOpenVMS: OpenVMS When your system absolutely has to runWhat is OpenVMS?: What is OpenVMS? The dominant force in the minicomputer revolution Rated as the #1 OS in the healthcare industry (1996) The first OS to run 24x365 (1985) and still the leader The first OS to scale from desktop to data center (1987) The first disaster tolerant OS (1988) and still the leader The first OS to ship CORBA, one of the first to ship DCE, and the first non-UNIX OS to ship POSIX and SPEC 11/70 The first OS to run 3-tier client/server and still the leader The second OS to run 64-bits The most highly scalable OS in the world today OpenVMS History - 20th Anniversary : OpenVMS History - 20th Anniversary 1976 Programmable Data Processors 16-Bit PDP 1977 Virtual Address eXtened 32-Bit VAX 1977 Aug Virtual Memory System v1.0 shipped on on VAX 11/780 hw /VMS sw 1983 First ship VMS Clusters VMS 4.0OpenVMS History - 20th Anniversary : OpenVMS History - 20th Anniversary 1986 First ship of single chip 32-Bit single chip microprocessor MicroVAX II with VMS v4.2 1988 VMS supports first Symmetric Multiprocessing Processing software for general purpose computing on VAX 6000s VMS v5.0 OpenVMS History - 20th Anniversary : OpenVMS History - 20th Anniversary 1990 Display list DECwindows GUI (VWS) is replaced by DECwindows Motif with VMS v5.2 1991 Disaster Tolerant Clusters 150 miles apart 1992 VMS becomes OpenVMS with Posix and Spec 11/70 compliance X/open XPG3-XPG/4 Branded just like Unix 1992 OpenVMS Alpha 64bit support first delivered in Feb OpenVMS History - 20th Anniversary : OpenVMS History - 20th Anniversary 1993 OpenVMS v5.5-2 (Oldest Y2K certified OS) 1994 OpenVMS v6.1 VAX 32-bit and Alpha 64-bit functional equivilency, cross architecture clustering 1995 OpenVMS v6.2 SCSI clusters 1995 OpenVMS / WNT Affinity Program Announced in May 1996 OpenVMS v7.0 Alpha 64-bit differenciators from VAX rollout OpenVMS History - 20th Anniversary : OpenVMS History - 20th Anniversary 1997 OpenVMS v7.1 Alpha 64-bit applications Oracle 7.3 VLM/VLDB 1997 OpenVMS WNT Affinity program Announced Wave 4 with ISG Navigator and over thirty products for Windows Interoperablity and Client / Server Support 1997 Disaster Tolerant Cluster 540 Miles, Memory Channel (100Mbyte/Sec) Clusters 1997 OpenVMS / Galaxy Software demonstrated virtual machine, OpenVMS Clusters in a box.You’re going to have to answer some questions:: You’re going to have to answer some questions: Is OpenVMS a good long-term strategic investment? Is OpenVMS also strategic to Oracle?OpenVMS Business Summary: OpenVMS Business Summary FY98 - $70M budget 500 person Engineering organization Oracle 12 engineers onsite $1.5M funding #1 Operating System in Healthcare* 23% growth 15% new business * based on 1997 CHIME studyOpenVMS Engineering plan for the next five years: OpenVMS Engineering plan for the next five years Deliver innovative, leading edge Affinity systems in partnership with customers into the 21st century Be the ultimate high-end for Windows NT Be a superb Database and Internet server Raise the bar for availability Raise the bar for scalabilityOpenVMS and Windows NT Strategy: OpenVMS and Windows NT Strategy OpenVMS as Continuous Computing Server Enterprise strength Bulletproof 24x365 Highly scalable Proven in the most demanding environments 64 bits/VLM Highly Available Internet Servers Memory Channel Clusters Galaxy OpenVMS & Windows NT Connectivity Applications Development environment Momentum PC volume economics 3 Tier C/S ADE Seamless Access to Data Affinity System Management Windows Gateway to OVMS Distributed TP “The seamless integration of OpenVMS with Windows NT will enable OpenVMS customers to combine the proven enterprise strengths of OpenVMS with the new foundation of Windows NT applications.” - Bill Gates Microsoft/Digital, August 2, 1995…and a few more questions: …and a few more questions Does OpenVMS have the mission critical functionality we need? Does OpenVMS scale? Where is OpenVMS going in the future? What is “excellence” in availability in 1997?: What is “excellence” in availability in 1997? Paris: May 5, 1996OpenVMS clusters offer a long list of unmatched availability features : OpenVMS clusters offer a long list of unmatched availability features Up to 96 nodes and 28 petabytes in a cluster Distributed lock manager Cluster file system Multi-site deployment Wide-area shadowing Automatic failover Automated volume shadowing On-line back-up Rolling upgrades for: Hardware Systems software Application softwareOpenVMS certainly scales well enough as a stand-alone system: OpenVMS certainly scales well enough as a stand-alone system Rdb Gains Index scan - 55X Indexed Data Retrieval - 146X 3-way Join - 142X 5-way Join - 216X Single system record holder - 14,227 tpm-C - April 1997 64 Bits for VLM, VLDBAnd OpenVMS scales REALLY well in clusters…and can be multi-site disaster-tolerant besides: And OpenVMS scales REALLY well in clusters…and can be multi-site disaster-tolerant besides Internet NYC Paramus Single logical data imageLeadership functionality:Disaster-tolerant computing: Leadership functionality: Disaster-tolerant computing System Disk Data Disk Data Disk Shared Disks GIGAswitch System Disk Data Disk Data Disk Shared Disks GIGAswitch T3 or FDDI or ATM Shadowed Disks Site “A” Site “B” Easy to manage - Single system view Operation Management Software provides management of configuration from either site Automatic transfer in seconds in event of disaster Integrated, tested, installed system solution - 5 yrs of experience OpenVMS Galaxy Technology: OpenVMS Galaxy Technology Brings multi-system performance, availability & scalability to a single system An extension/evolution of current OpenVMS arch S/W arch that adapts to the way you do computing partitions a single machine into multiple SMP nodes environment is adaptive and dynamic multiple instances of OS execute in a single computer Addresses limitations in current system archs: clusters: lack of shared memory and interconnect SMP: S/W scalability and single system view MPP: limited applications, management Digital Proprietary and ConfidentialOpenVMS Galaxy Technology (cont): OpenVMS Galaxy Technology (cont) Consists of two components: APMP (Adaptive Partitioned Multi-Processing) model Galaxy Softwared Architecture to exploit APMP 4 Modes of Operation: Shared Nothing: partition CPU’s for dedicated apps Shared Partial: shared memory for VLDB Shared Everything: shared mem & cluster interconnect Adaptive Dynamic Resource Partitioning: migration resiliency Capabilities: 1-256 CPUs 300,000+ to 1,000,000 design goal dynamice resource allocations hot swap of CPU, mem, I/O Digital Proprietary and ConfidentialNeed bigger - Galaxy: Need bigger - Galaxy Node A I/O Node B I/O Node C I/O Shared Memory Digital ConfidentialGalaxy Software Architecture: Galaxy Software Architecture Node A I/O Node B I/O Node C I/O Shared Memory Digital Proprietary and ConfidentialGalaxy Software Architecture withAdaptive Dynamic Resource Partioning: Galaxy Software Architecture with Adaptive Dynamic Resource Partioning Node A I/O Node B I/O Node C I/O Shared Memory Digital Proprietary and ConfidentialGalaxy Implications: Galaxy Implications Shared memory Very large database cache Node reboot into “hot” cache DLM cache now local vs distributed Shared memory cluster interconnect SMP Improved SMP scaling Improved SMP affinity CPU Migration Partitioning of CPUs I/O Multiple primaries for INT Distribute INT to secondaries Can partition I/O workload Highly scalable I/O subsystem Reliability / Availability Hotswap (CPU, Memory, I/O) OS will fail out only the necessary subsystem Software failure will not compromise entire system OS Resilliency (dynamic reallocation or resources) Digital Proprietary and ConfidentialSlide27: www.digital.com www.openvms.digital.com www.montagar.com/dfwlug/