Presented by Terry C. Shannon, Shannon Knows HPC Director, T.C. Shannon and Associates LLP...

download Presented by Terry C. Shannon, Shannon Knows HPC Director, T.C. Shannon and Associates LLP Publisher, Shannon Knows HPC

If you can't read please download the document

description

The Fine Print and the Usual Caveats  Sad but true, the solicitors require me to say all these things!  SKHPC is an independent publication which is not authorized by, affiliated with, vetted by, or endorsed by HP.  SKHPC’s publisher is not, and never has been, an employee of DEC, CPQ, HPQ, or any other IT vendor.  Hewlett-Packard, the HP Logo, and HPQ are trademarks of HP. Products cited in this presentation may be trademarks and/or registered trademarks of their respective companies, or of third-party companies. Some of this material is contingent on the successful and timely implementation and execution of the HP Acquisition. Future strategic or product-specific decisions made by HP could affect the content of this presentationSome of this material is contingent on the successful and timely implementation and execution of the HP Acquisition. Future strategic or product-specific decisions made by HP could affect the content of this presentation  This material reflects SKHPC’s opinion and should be construed as an explanation, not a defense, of Compaq’s IPF Consolidation, HP’s acquisition of Compaq, or HP’s product roadmaps, strategies, and organizational changes.  Caveat lector! All the material in this presentation is believed to be accurate, however its accuracy cannot be guaranteed.  SKHPC does not deal in “Probability Factors”… but our twenty-year accuracy record exceeds 90 percent.  Trust but verify: This presentation contains no information obtained under nondisclosure. SKHPC strongly suggests that you obtain proprietary or futures-related information from HP before making purchasing decisions. Since it’s hard to predict, especially the future, please do not rely solely on commentary from any analyst without getting a second opinion!  Please excuse my American English (a contradiction in terms!)  All material Copyright © 2003 by Terry C. Shannon and is freely redistributable (with attribution) to attendees.

Transcript of Presented by Terry C. Shannon, Shannon Knows HPC Director, T.C. Shannon and Associates LLP...

Presented by Terry C. Shannon, Shannon Knows HPC Director, T.C. Shannon and Associates LLP Publisher, Shannon Knows HPC Presented by Terry C. Shannon, Shannon Knows HPC Director, T.C. Shannon and Associates LLP Publisher, Shannon Knows HPC The New HP: 15 Months Later The Fine Print and the Usual Caveats Sad but true, the solicitors require me to say all these things! SKHPC is an independent publication which is not authorized by, affiliated with, vetted by, or endorsed by HP. SKHPCs publisher is not, and never has been, an employee of DEC, CPQ, HPQ, or any other IT vendor. Hewlett-Packard, the HP Logo, and HPQ are trademarks of HP. Products cited in this presentation may be trademarks and/or registered trademarks of their respective companies, or of third-party companies. Some of this material is contingent on the successful and timely implementation and execution of the HP Acquisition. Future strategic or product-specific decisions made by HP could affect the content of this presentationSome of this material is contingent on the successful and timely implementation and execution of the HP Acquisition. Future strategic or product-specific decisions made by HP could affect the content of this presentation This material reflects SKHPCs opinion and should be construed as an explanation, not a defense, of Compaqs IPF Consolidation, HPs acquisition of Compaq, or HPs product roadmaps, strategies, and organizational changes. Caveat lector! All the material in this presentation is believed to be accurate, however its accuracy cannot be guaranteed. SKHPC does not deal in Probability Factors but our twenty-year accuracy record exceeds 90 percent. Trust but verify: This presentation contains no information obtained under nondisclosure. SKHPC strongly suggests that you obtain proprietary or futures-related information from HP before making purchasing decisions. Since its hard to predict, especially the future, please do not rely solely on commentary from any analyst without getting a second opinion! Please excuse my American English (a contradiction in terms!) All material Copyright 2003 by Terry C. Shannon and is freely redistributable (with attribution) to attendees. Some Words of Thanks Its a pleasure being here and addressing this group! Special Thanks are in order for Hewlett Packard The OpenVMS Ambassadors All of you who took time out of your schedule to attend this event Rich Marcello, Mark Gorham, Mary Ellen Fortier, Sue Skonetski The conference producers Speakers, keynoters, and HP employees And the folks behind the curtain who made this event a great success PS - Dont forget to fill out your questionnaires before you leave! It Isnt Easy Being the Keynote for a VMS TUD Meeting But I Guess Ill Drop In and Give It a Try! My Name is Terry, and I am a VMS Veteran I first picked up VMS 3.4 over 20 years ago as a VAX 11/750 system manager. It was a highly addictive experience. Less than a year later, I wrote Introduction to VAX/VMS and well over 150 VMS-related articles, papers, and presentations. Directed IDC DEC Advisory Service, helped form Illuminata, Inc. Launched ten years ago Launched Shannon Knows DEC/Compaq/HPC ten years ago Aided the VMS effort via customer roadshows, briefings, user group meetings, presentations, and CIO breakfasts worldwide. And I think the best may be yet to come VMS comes to Itanium (be careful what you wish for) VMS remains the GOLD STANDARD for clusters Management enhanced by OpenView VMS expertise will advance UDC and Adaptive Infrastructure Security and disaster tolerance issues: Security and disaster tolerance issues: not if, but when VMS is a springboard to ubiquitous mission-critical computing June 25, 2001: Compaq EOLs Alpha History will prove that Compaq made the right decision when it decided to cancel EV8 and transition, over time, all of its enterprise systems from Alpha to Intels Itanium Architecture. And yes, you may quote me on this Make my day, skeptics! SPECint95 Relative Performance:Integer (SPECint95-peak) MHz u EV7 IA-64 (Merced ) MHz EV8 Alpha Roadmap/Performance Curve Circa 1998 EV68 Alpha will maintain its performance leadership 64-bit capability on WNT well before it debuts on Merced, er, McKinley EV78 CPU Evolution Changed the Curve EPIC Superscalar RISC ~ 2 instructions/cycle performance time RISC < 1 instruction/cycle.3 ins/cycle next generation CISC 20-30% increase per year due to advances in underlying semiconductor technology 1 micron - >.5 micron -->.35 micron -->.25 micron -->.18 micron -->.13 micron Itanium you are here IA-32 IA-64 IA-64 Itanium * IA-32 expands leading edge performance for volume segments IA-64 enhances IA with world class performance for workstations and servers and full compatibility A Unified CPU Architecture 32 Bits and 64 Bits Pentium Pro Pentium II Pentium III Pentium II Xeon.... Foster Cascades Itanium 2 MadisonIA64-Perf Hondo Montecito Future IA32 Gallatin Pentium-M Future IA Intel Architecture Roadmap We are Here A CPU Micro-Architecture Comparison Intel Itanium 2 McKinley 6.4 GB/s 8 1 GHz System Bus Bandwidth On-die Cache On-die Registers Execution Units Core Frequency Issue Ports 328 Registers 6 Instructions / Cycle L3 = 3 MB, L2 = 256K; L1 = 16K + 16K Instructions / Clk 6 Integer, 3 Branch 2 FP, 1 SIMD 2 Load, 2 Store RISC Architecture *8 MB L2 External Cache Sun UltraSparc* III 96K L1* 14 2 Integer 1 Branch 2 FP/VIS 1 Load/Store 900 MHz Registers 4 instructions / Cycle 2.4 GB/s EPIC ArchitectureSource:The SPARC Architecture Manual (Prentice Hall)www.sun.com After the Alphacide, Compaq and HP Still Needed Something September 4, 2001 Compaq and HP Get an Urge to Merge An Acquisition, not a Merger New name: Hewlett Packard ~$87B USD Company ~135K employees after ~15K layoff NSK, OpenVMS are safe Tru64 merges with HP-UX The New HP Summarized: Promises Made, Promises Kept HP has delivered on virtually every promise it made on 7 May 2002 However, there have been a few pleasant surprises along the way And one unpleasant one 3FQ03 Results Agenda Hardware OSes (Focus on VMS) Alpha RetainTrust Program UDC and Adaptive Infrastructure HP 15 Months Post Merger HP Said It Would Reduce structural costs Implement aggressive go-to-market programs Roll out product roadmaps and customer transition plans Continue to grow IPG and HPS, and improve the PSG and ESG Drive innovation Transform the industry And What HP Has Done Met or exceeded all of our cost savings goals Met or exceeded all of our cost savings goals Implemented aggressive go-to-market initiatives Implemented aggressive go-to-market initiatives completed sales force alignment with a single compensation plan completed sales force alignment with a single compensation plan took targeted pricing actions to improve competitive position/share took targeted pricing actions to improve competitive position/share launched new brand and demand generation advertising campaigns launched new brand and demand generation advertising campaigns Executed product roadmaps and transition plans Executed product roadmaps and transition plans Business groups newly reorganized Business groups newly reorganized Grew IPG and extended market leadership Grew IPG and extended market leadership Maintained strong operating margin in HPS Maintained strong operating margin in HPS Returned PSG to profitability in Q1 Returned PSG to profitability in Q1 Improved ESGs loss by 75% from Q3 to Q1 Improved ESGs loss by 75% from Q3 to Q1 Added >1,400 patents during last half of fiscal 02 to portfolio of >17,000 patents worldwide Added >1,400 patents during last half of fiscal 02 to portfolio of >17,000 patents worldwide Fostered fastest rate of innovation in HP history Fostered fastest rate of innovation in HP history You want it? HP has it NOW! Workgroup Computing Supercomputing Enterprise Computing Blades-1-2-4way AlphaServer GS and ES Superdome SC supercomputer rp7410 rp8400 rp5470es45 Interconnect Fabrics hyperfabric2Gb/ethernetquadrics elan Linux supercomputer HP Superdome: Built for the Long Haul With Great Investment Protection Today & Tomorrow 2h 2001mid 20022h 2003 PA CPUs 1 TB RAM hp-ux online replacement and deletion of cell boards PA MHz 64 CPUs 256GB RAM new I/O cards Itanium* 64 CPUs 512 GB RAM hp-ux, Windows, Linux PCI-x online addition of cell boards cell local memory PA percent performance increase to 389k TPC-C Real Soon Now Itanium * 128 CPUs 1 TB RAM hp-ux, Windows, Linux, VMS in the future 1h 2004 *based on the Itanium processor available at that point in time A Marvel-ous EV7 AlphaServer GS1280 HP Multi-OS Hardware, Circa ~2007 Note: This slide is based on analysis of current products, announced roadmaps, and public data. As such, it is hypothetical and conjectural. Goal: develop and deliver The Mother of all Enterprise Servers SKHPC assessment of product development and attributes Produced by ~50-person IPF development section as part of the High Performance Systems Lab (HPSL) in Hardware Systems Technology Division (HSTD). Distributed in SHR, MRO, Colorado, Richardson. Advanced development team focused on roadmap linkage, definition, architecture, topology, performance work, as well as base technology assessment and proofs of concept Implementation team to deliver an SPU (system processing unit) consistent with current HSTD development practices including boards, signal integrity, mechanical, power, thermal, utilities, etc. Team will leverage existing HP ASICs, CPUs, diagnostics and IO options. Design and appearance: Superdome-centric with Alpha attributes Result: Post-Superdome with additional performance and scalability The first, true, open-systems alternative to proprietary mainframe and parallel cluster technology. And, a new ability to attack the Main Rival of : IBM, of course! (Sun isnt likely to be a factor in 2007) The Ideal Server in 2007: SKHPCs Vision No limit to Moores law so CPU speeds in> 6 GHz region CPU counts to ~256 processors which allows HP deliver SC-like systems in a box partitioned all the way down to individual processes using the continuum of partitioning technologies An order of magnitude improvement in availability with proactive diagnosis tools that anticipate possible failures and automatically correct through re-configuration or early replacement of failing components Incredibly large cache and memory sizes allow you to run almost any app in memory at the fastest possible speed TPC-C numbers of several million (conservative estimate) HPTC performance so high that HP can model virtually any scientific phenomenon Easily clustered, interoperable, and a component of the HP Utility Datacenter (Built on inexpensive industry standard components and able to run all HP OSes and the full set of industry applications HPs Operating Systems HP-UX (youve heard plenty already!) Tru64 UNIX (portions already merged into HP-UX) Linux (this is the one to watch!) Windows (too much has been said about this!) OpenVMS (Dead Meat? I dont think so) And Now The $800,000 (or 64-bit) Question: Is HP Serious About the 64-bit OVMS World? VMS is NOT Going Away: A Plethora of Proof Points Decision to maintain VMS was made very early in Clean Room Hundreds of thousands of systems, millions of users High margins and stellar synergy with HP Services 200 dedicated OpenVMS Ambassadors worldwide VMS: far more successful in 25 years than Ive been in 50 years The gold standard for clustering and HP has it now! All RAS, no FUD-laden futures Underpinnings for Galaxy APMP, future adaptive systems VMS on IPF: It LIVES! A bright present, and an even brighter future for VMS Increasing interoperability and commonality with Unix and Linux Ironclad security and immunity to worms, viruses, and hackers A return to growth mode entirely possible! VMS: Carly Likes it, Too!. OpenVMS: The Song Remains the Same Moving forward: Why Stay With Alpha/VMS? What Alpha/VMS Customers are Asking: Show Us the Business Value Need compelling business reason (e.g. cost savings, improved ROI) beyond the migration that results in competitive edge in the marketplace Dont force us to transition for the sake of technology Minimize cost of transition for entire IT environment while increasing total TCO savings Demonstrate why the next generation IT move with HP is better for my business than a move to IBM or Sun, if theyre still a player in the business! Help Us Take Away the Risk Ensure application availability Absolutely no unplanned downtime during introduction of new environment Minimize impact to IT support staff through training on new environment and how to transition guide New environment becomes a superset of the existing environment to solve next generation business problems Prove HPs commitment by delivering on roadmaps; no forced transition; gradual transition with minimum risk The Alpha RetainTrust Program HP Will Extend Alpha Viability and Value by Building upon 25 years of solid experience Creating and extending the gold standard of cluster technology Leveraging experience in 10-year chip lifecycle transitions Delivering virtually unprecedented levels of guarantees to our Alpha customers With our partners, providing leadership in key market segments The Bottom Line: Alpha remains a safe bet for Alpha customers today! A Suite of Investment Protection Choices AlphaServer Customer Assurance program available now who purchase new AlphaServer systems expecting to transition to the Itanium architecture Guarantee available for customers who purchase new AlphaServer systems expecting to transition to the Itanium architecture Sets the standard for customer satisfaction through, for example, money-back guarantees To register: visitenterprise/cap_reg.html Software trade-in proposal stage , to ease the transition of platform changes Special offer, planned for Q2 CY03 availability, to ease the transition of platform changes VAX/Alpha & HP-UX customers continue with existing policies VAX/Alpha customers transitioning to HP-UX assume HP-UX policies Tru64 UNIX/AlphaServer HSG80 trade-in program under development to HP-UX Eases storage transition to HP-UX Goal: Worldwide program for Q3 CY03 availability For more info, contact Rich Marcello or Mary Ellen Fortier Transition Lifecycle: A Systematic Approach Aware- ness (now) Design (1 yr 6 months prior to transition ) Plan (1-3 years prior to transition) Assessment of business drivers and product roadmaps Selection of HP destination servers and timeframe Detailed planning Resource scoping Implementation plan Development of transition configuration (BOM, layouts) Provisioning of servers and storage, and installation of OE and application software Data migration and validation Deployment Engage with HP to understand HPs strategy for moving forward (Alpha RetainTrust ) HP account team Seminars, users groups, advisory councils Web portals and web- based training Post-transition activities Backup system strategy Performance monitoring & tuning Implement (2 months prior to transition ) Manage (ongoing after transition) Alpha RetainTrust Program Summary Provides you with: , ensuring long- term sales and support HPs commitment to pre-merger product roadmaps, ensuring long- term sales and support while continuing to support the AlphaServer platform ISV enthusiasm about the transition to Itanium while continuing to support the AlphaServer platform for a smooth transition Investment in transition tools, services and programs for a smooth transition A A transition to the Itanium architecture on your timetable for which you have come to depend Continued delivery of the capabilities and solutions for which you have come to depend More than 200 ISVs have committed to port over 400 apps to OpenVMS on Itanium, ~100 ISVs are expected to commit within the next few months. OpenVMS engineers are successfully recompiling partner and customer apps and have had little difficulty with the Alpha to Itanium ports. More than 200 ISVs have committed to port well over 400 apps to OpenVMS on Itanium. Among those who have elected to port are: Key industry and horizontal partners Developers responsible for hundreds of completer apps. ~100 ISVs are expected to commit within the next few months. Additional partners are committing to the Itanium platform as each OpenVMS porting milestone is achieved. Training and support programs are in place to provide ISVs with access to OpenVMS online or at porting centers. The Version 2 SDK will be available to ISVs in less than 6 months. OpenVMS engineers are successfully recompiling partner and customer apps on a test basis and have had little difficulty with the Alpha to Itanium ports. The porting center in Nashua, NH is running at full capacity and there is a substantial waiting list for access to porting hardware. HP OpenVMS ISVs Trust the IPF Port You Want Mixed Alpha/IPF Clusters? Or How About VAX/Alpha/IPF Clusters? HP OpenVMS - Better in the New HP! Faster / better / cheaper Itanium solutions Reduced development costs Tighter Integration with OpenView product Increased HP Classic Support Better ISV Leverage Enhanced Printer Support Enterprise Customer Mindset Better Marketing Ensured Future New growth opportunities More customer confidence (per SKHPC survey) More customer confidence (per SKHPC survey) June 25, 2001 HP Delivers on Its Commitments! Roadmap published 6/25/01 remains unchanged Delivered on schedule HP is delivering promised new systems on schedule HP AlphaServer GS EV7 (8-64p) HP AlphaServer ES EV7 (2-8p) HP AlphaServer DS 15 EV79 HP AlphaServer ongoing sales and support EV79 HP AlphaServer Roadmap Sales at least until 2006, support at least until 2011 Deliver HP AlphaServer systems according to published roadmap, including EV7 and EV79-based systems Protect customer investment through Alpha RetainTrust program Flexible upgrade paths, allowing customers to move when ready Birds (and lemmings and salmon) migrate, customers transition NO SHOTGUN MIGRATIONS!!!! HP OpenVMS - Better in the New HP! Faster / better / cheaper Itanium solutions Reduced development costs Tighter Integration with OpenView product Increased HP Classic Support Better ISV Leverage Enhanced Printer Support Enterprise Customer Mindset Better Marketing Ensured Future New growth opportunities More customer confidence (per SKHPC survey) More customer confidence (per SKHPC survey) Putting It All Together With AI and UDC Those of you who have seen my pre-merger Server Utility presentations will find the following several slides to be somewhat familiar Then Things Start to Get Interesting This works, today HP has a product in the lab and at customer sites Seen any self-healing ELIZA lately? Seen any Sun N1 goodies (other than advertisements? Unlike Compaq, HP is also virtualizing networking Servers: From Futility to Utility What goes around, comes around Mainframe Minicomputers Client Server Clustering ?Consolidation.. and server schizophrenia HPs RX for server schizophrenia Appliances - simple, fixed function Ultra Scalable Systems - all resources treated as one A Son of SAN Server Utility strategy UDC Precursor: ENSA Storage Virtualization RAIDArray Switched Fabric Fibre Channel Tape Controller Tape library Snapshot Pool of Computes Pool of Storage ENSA SAN Server Partitio n Server Partition Storage Server Partition Storage Server Partition Compaqs Server Utility: ENSA Extended Son of Storage Area Networks Synergy with ENSA Storage Utility model Flexible approach to server resource management System instances are allocated from a common pool of distributed resources Resources are dynamically reallocated to respond to changing needs Combining Virtual Storage and Virtual Servers Server Utility Compute Pool Storage Utility Storage Pool / /usr /var/... / /usr /var/... / /usr /var/... / /usr /var/... CPU And Managing Them With Galactic Techniques Memory & I/O CPUs CPUs Instance A Instance B Instance C CPUs Batch Jobs Mail Backups Order Matching Trade Processing Surveillance Data W.H. Data Mining Memory & I/O CPUs OpenVMS Galaxy APMP at the Stock Exchange Move the capacity, not the workload Going Beyond Galaxy and Beyond OpenVMS, too CPU Free CPU Pool SMP Instances CPU Cluster Standby Node Business as usual with primary and standby nodes Failover CPU Cluster X CPU failure invokes failover services CPU Cluster CPU X Failover node auto-magically grows to full capacity CPU Cluster Standby Node Excess resources deallocated to free and repair pools Repair Pool Server Utility Feature Summary Vastly reduced management complexity and cost Server consolidation on steroids Disparate, distributed resources managed as a whole Ability to respond rapidly to changing business needs Shift capacity (not workload) to where its needed React to the unexpected via a just-in-time growth model Quickly increase capacity or deploy new systems Nonstop boot-once operation Continuous availability at popular prices And the underlying architecture is nearing release... OpenVMS, Tru64 UNIX, NSK, Windows Datacenter, and Linux simultaneously Compaqs Future IPF-Inside Enterprise Server Managed system partitions across racks & modules. Fabric-based, can run OpenVMS, Tru64 UNIX, NSK, Windows Datacenter, and Linux simultaneously Compute Nodes I/O Devices Data Storage Rack Servers Compute Nodes Pools of CPUs & pools of storage Collection of OS and apps images Dynamic assignment and reassignment of resources Load Balancing Within Partitions CRM Solution Customer 1 Web servers Customer 1 Database Customer 2 Database Exchange Servers File Server Customer 2 Web servers ERP Solution Dynamic Provisioning Utility Pricing While Compaq Wanted to Virtualize Disks and CPUs, HP Decided to Go A Bit Farther Deal with all of the problems of todays data centers Extend Compaqs Adaptive Infrastructure strategy Employ all key HP Oses, including OpenVMS Deploy future IT centers as Utility Data Centers Wire your data center once Virtualize all of your resources and assets Let management software do all the heavy lifting And the network will be the system OpenVMS will play a strong role in UDC and AE IT as a utility was first envisioned by Ken Olsen nearly 20 years ago. Why UDC and Adaptive Enterprise? Ability to Adapt Quickly Business challenges Improve business performance, quality and ROI, while reducing costs Minimize risk associated with change Drive new business models and direction Shorten time-to-market Enable mergers, acquisitions and divestitures IT IT imperatives Link business and IT Reduce costs, ensure stability and flexibility Reduce complexity Optimize assets today and tomorrow Extend value and reach of the enterprise Reduce headcount Increased Volume of Change The Downsides of Todays Data Centers Very Expensive to Operate Requires manual, labor-intensive deployments for even minor changes to IT environments Inefficient asset utilization because of lack of data center-wide load balancing capability Development, support and use of home-grown management applications is cumbersome Inflexible and Complex Vendor provides a fixed, inflexible architecture to each customer regardless of specific needs Highly complex overall architecture is required to accommodate each users and apps needs Difficult to scale because of evolutionary growth or react rapidly to spikes in demand Error Prone and Unreliable Human intervention accompanies every change request and implementation Introduction of new infrastructures is invariably slow, costly, and disruptive. No data center-wide high availability threatens critical or irreplaceable data. No integrated management view of all services and operational environment HP Adaptive Enterprise in One Slide business strategy and processes HP adaptive enterprise applications dynamic resource optimization continuous and secure operations automated and intelligent management foundation for the future What is HPs Utility Data Center? hp utility data center - virtualized pools of resource for instant ignition - failover protection and data replication to protect servers, storage and network - wire-once fabric - utility controller software for service definition and creation New applications and systems can be ignited within minutes Server, storage and network utilization approaches 100% Resources are virtualized and optimize themselves to meet your service level objectives Administrative and operational overhead is minimized storage virtualization network virtualization internet server pool storage pool NAS pool load balancer pool firewall pool switching pool utility controller server virtualization HP Utility Data Center Components Virtual Server Pools Heterogeneous server environments HP servers optimized for UDC Protect your current investments Virtual Network Pools Standards-based VLANs Flexible and robust network infrastructure Virtual Storage Pools HP XP and EVA storage offer flexible network-based virtualization Integration with OpenView for storage management EMC Symmetrix Utility Controller Software Manages service templates Integrates with HP software: resource, workload and failure mgt. OpenView processing elements storage elements hp-ux OpenVMS windows linux utility fabric networking elements emc hp utility data center hp utility controller software HP consulting and integration services Cisco Procurve eva hp-xp HP UDC: Improving Asset Utilization storage virtualization network virtualization switching pool utility controller server virtualization internet load balancer pool firewall pool NAS pool server pool storage pool switching pool Wire it up just once network, storage, and server components wired once Virtualize asset pool All components can be allocated and reallocated Easily reconfigure simple user interface allows administrators to architect and activate new systems using available resources Adaptive management solution enabling virtual provisioning of application environments to optimize asset utility Operations center rack Utility controller (Mgmt rack) Backup rack Storage array Fabric rack HP UDC at the Palo Alto Research Labs. 12/02 HPs Utility Data Center in Palo Alto, USA The Business Case for UDC upgrading & migration 20% 40% reducing costs upgrade & migration economies usage metering 5% 30% more accurate charge-back and billing metering economies security 20% 30% self adaptive 80% 100% reducing costs economies provisioning & operational economies deployment 30% 80% capacity planning 5% 10% improved asset utilization asset utilization economies higher server and storage utilization 5% 40% SKHPCs Top 5 OpenVMS Predictions Here are some Key OpenVMS Predictions Many of these will be reality by the next TUD Prediction Number 1 First VMS Industry Standard 64 SDK Today OpenVMS on IPF, aka OpenVMS Industry Standard 64 OpenVMS on an industry-standard platform? Be careful what you wish for, sometimes you get it! Targets include Targets include Key ISVs, Partners, Early Adopter Contents of OpenVMS V8.0 Mako Monitor Utility DECnet Phase IV, TCP/IP Development Tools Cross Linker, Librarian Compilers C, C++, BLISS, FORTRAN, IMACRO Prediction Number 2 Second VMS/IPF SDK before end of year Target Audience: Target Audience: Key ISVs, Partners, Early Adopters Contents of OpenVMS V8.1 Jaws Limited cluster functionality (4 nodes) Native Compilers C, C++, BLISS, FORTRAN, IMACRO, Pascal, BASIC, COBOL Additional Layered Products Networks, Data Serving Security eBusiness Integration Application Development Prediction Number 3 Production VMS Industry Standard 64 in 2004 Candidate Components for OpenVMS 8.2 Topaz Enhancements System & I/O Performance Enhancements Enhancements Cluster Interconnect Enhancements including Fibre/SAN enhanced support including Enhancements Disaster Tolerance Enhancements Continued Continued Security Enhancements More More updates for e-Business and Integration release, including Clusters Alpha Compatibility with OpenVMS Itanium release, including Clusters More features More UNIX Portability features Prediction Number 4 Continued VMS/Itanium ISV Attraction New ports announced almost daily VMS/UNIX interoperability aids porting effort HP is increasing focus on Itanium ports Former customers returning to VMS Success of IPF port is the causative factor Windows penetration of enterprise is slow Clustering, RAS, Disaster Tolerance are boosting VMS acceptance Potential licensing changes may reduce entry price for VMS (variation on HP-UX model) Prediction Number 5 VMS/Itanium will proliferate Gating factor: low-cost Itanium desktops Not a near-term development SKHPC estimates 4-5 years for low-cost systems VMS on Itanium Laptops? Definitely, when the price is right (desktops much sooner) VMS on non-HP Itanium systems? Almost certainly Any Itanium system that plays by the rules is a potential VMS host VMS port to another 64-bit architecture? Not currently planned But not beyond the realm of possibility IT Will Be a Two Horse Race Well, he always ran best In the mud.