Как выбрать оптимальную серверную архитектуру для создания высокоэффективных ЦОД
Upcoming SlideShare
Loading in...5
×

Like this? Share it with your network

Share

Как выбрать оптимальную серверную архитектуру для создания высокоэффективных ЦОД

  • 2,161 views
Uploaded on

 

More in: Technology
  • Full Name Full Name Comment goes here.
    Are you sure you want to
    Your message goes here
    Be the first to comment
    Be the first to like this
No Downloads

Views

Total Views
2,161
On Slideshare
2,153
From Embeds
8
Number of Embeds
2

Actions

Shares
Downloads
16
Comments
0
Likes
0

Embeds 8

http://www.pcweek.ua 7
http://www.slideshare.net 1

Report content

Flagged as inappropriate Flag as inappropriate
Flag as inappropriate

Select your reason for flagging this presentation as inappropriate.

Cancel
    No notes for slide
  • Slide Purpose: To discuss where and why IT is spending their limited budget. Get the customer to reflect on their “big picture” IT strategy (how do they utilize their budget) (illustrate on the next 2 slides .. What are the alternatives) IT is increasingly under pressure to innovate!!! Different parts of your organization are clamoring for different needs Internal Users: More performance and Faster Applications CEO – calling for Growth .. Driven by new services, new customers, new markets CFO – calling for lower costs .. Customers .. Calling for new services, faster However, the majority of IT’s annual budget spend is focused on maintaining the capabilities already in place . (talk to the gartner findings on where/how IT is spending their money. (relfect) Where Are You? … Managing Data Center Health is about balancing investment issues in light of your business goals (delivering performance and capability)  new technology deployment vs (managing resources – budget, space, power, people)  resource management The real benefits of refresh can be gained when you develop an IT strategy that encompasses business needs, available resources, technology roadmap and then optimizes the benefit/cost equation with refresh as an central element Prioritize What’s Most Critical Evolve a Strategy to Maximize Value & Capability over time In many ways … dc refresh is like maintaining your personal health, there are many considerations to balance and regular check-ups and education are constantly required. … And like individuals, each DataCenter and business is unique and prioritizes requirements differently and available assets There is a common approach that every IT department uses … technology refresh of their hardware and software environment … the constant questions are when, how and why
  • Key Message: Highlight Intel IT’s investment strategy in we are being extremely proactive. We’re not just talking about it, but doing it.
  • Key Message: provide a starting point for the dicscussion on how fast. Intel IT recently conducted a detailed analysis to evaluate how fast was right for them They looked at a variety of aspects of TCO and determined hat 4yrs was maximize ROI taking into account all aspects. This is consistent to other 3 rd party studies about typical refresh Recently I heard the industry average (before) dual-core was 4.5yrs .. Many analysts have stated that with virtualization, the transition to multi-core, and the increasing importance of energy efficiency, the average refresh rate has increased
  • Slide Purpose: Communicate the End Benefit and Alternative approaches to Refresh of Older Servers (Build slide) The culmination of the opportunity highlighted in the previous slides are the benefits and approaches you can take in refresh Let’s take the base case of 184 servers (could be any number of servers) .. If you replace each server 1:1 .. You can see a dramatic boost in infrastructure performance due to the per server performance gains available.. However, because the power is lower per server, you can do this with the same data center design and footprint that you have today (no retro-fit of expansion) .. And you can capture a moderate savings in your power bill or about 18% annually Another approach (used by many) is to replace your aging servers with fewer new. If you kept performance of your infrastructure constant (before / after), 9:1 consolidation is possible and a dramatic cost reduction can be gained in areas that depend on server count (like energy costs, software licensing and maintenance fees – among others) .. We estimate that just looking at OS and energy savings, the cost of the new servers could be paid back within 8 months from these savings.
  • So Powerful that Even a Dual core refresh makes financial sense with the new Intel microarchitecture Ideal For Space Constrained Businesses or Replacing Servers Coming Off 3-year Lease Terms or due for extended warranty protection. In this situation the payback is not as fast but for servers coming off lease and for businesses looking to be more competitive, address space constraints and reduce operating costs .. A dual core refresh after only 3 years makes good financial and business sense. Extended warranty protection can range from 800-1200 per server for another 2 years. That cost can be up to 20% the value of a new server
  • Key Message: Innovation continues, and our next step will be a new platform complete with several new innovations, which extends the leadership IT value prop we have delivered the past few years. Main features: Processor: 4-wide arch (just like Intel Core uarch), 33% more efficient than AMD Opteron (more instruction per clock), more parallelism (33% more micro-ops over 45nm Penryn), enhanced algorithms and branch prediction capabilities Hyper-threading – good for highly threaded apps such as databases, multi-media, search engines Turbo Boost – raises core clock speed when needed above rated speed (performance on demand) Integrated Memory Controller: 3 ch DDR3 memory per socket, up to 144GB (8 GB DIMMs) and 32 GB/s High Bandwidth QuickPath Interconnect: Up to 25.6GB/s per link Performance: Up to 3.5x the bandwidth for technical computing† Up to 2.25x the performance for enterprise computing† 2x software threads and performance boost on demand Energy Efficiency: Highest system level performance/watt† Lower system idle power reduces IT costs Dynamically turns cores on/off to meet performance needs Virtualization: Expand existing pools with Intel VT FlexMigration Unique capabilities for live VM migration over network Highest 2S virtualization scores mean more VMs per server†
  • Intel Turbo Boost Technology increases performance of both multi-threaded and single threaded workloads. Intel Turbo Boost Technology is activated when the Operating System (OS) requests the highest processor performance state (P0). The maximum frequency of Intel® Turbo Boost Technology is dependent on the number of active cores. The amount of time the processor spends in the Intel Turbo Boost Technology state depends on the workload and operating environment, providing the performance you need, when and where you need it. Any of the following can set the upper limit of Intel Turbo Boost Technology on a given workload: Number of active cores Estimated current consumption Estimated power consumption Processor temperature When the processor is operating below these limits and the user's workload demands additional performance, the processor frequency will dynamically increase by 133 MHz on short and regular intervals until the upper limit is met or the maximum possible upside for the number of active cores is reached. Conversely, when any of the limits are reached or exceeded, the processor frequency will automatically decrease by 133 MHz until the processor is again operating within its limits. Example for 4C Turbo operation: A database query, which doesn't use any of the processor’s floating point silicon, so it can take advantage of that thermal headroom and increase the frequency of all 4 cores.
  • Slide objective: Because of constant advancements in server technology over time as we (Intel) pursue the promise and reality of Moore’s law. The key point to make is that older 2S servers installed in your infrastructure are consuming too much power, space and requires too much maintenance / patching. Gains across a range of industry standard workloads Up to 3.5x the bandwidth for technical computing Up to 2.25x performance for enterprise computing
  • New Intel Xeon Processor 7500 (Nehalem-EX) – launching in Q1’10 is best enabler of IT business value we’ve seen in years up to 3x performance gains over existing up to 9x the bandwidth of up to 18x the performance of the most powerful single core servers installed today. 2 ways to use this performance Energy Efficient Refresh: Use 18:1 server replacement or virtualization and achieve a rapid 10 month payback with up to 90% lower operating costs Performance Refresh Use 1:1 server replacement to deliver nearly 18x performance gains within an existing data center design Or Deploy Virtualization: Boost utilization and flexibility of your infrastructure by capitalizing on these performance and energy efficiency gains
  • Xeon 5500 brought tremendous generation-to-generation performance jump NHM-EX will bring even higher performance leaps…biggest in Xeon product history Memory bandwidth measured as 9X (made public at May ‘09 NHM-EX press event) Database measured perf is up to 3X (to be made public at Sept ‘09 IDF) Integer and floating point perf #s for NHM-EX are not yet public. But we expect better than EP gen-to gen perf jump Database perf increase is vs standard Dunnington-Caneland platform (comparison vs a Dunnington system using other 3 rd party chipset [ie not Clarksboro IOH] would be different)
  • The following assumes your customer is open to purchase, so what to buy? There’s a range of different Xeon® 5500 SKUs available, segmented into three main usage bands – basic, standard, and advanced. Historically, we’ve typically differentiated SKUs primarily by frequency, but the advanced capabilities in the Xeon® 5500 platforms have enabled us to go beyond just frequency to include CPU L3 cache, QPI speed, memory speed, HT, and Turbo. As you spend more, you get more capabilities such as access to higher platform bandwidth, faster memory speeds, and advanced features like HT and Turbo Boost. Most people buy in the standard range as these SKUs offer a great balance of price, performance, and advanced features. If you’re focused on price, you still outstanding performance with the basic SKUs relative to existing products.
  • Here’s the performance stacks up with the different segments. As you can see, the “performance shelves” between segments provide a tremendous boost in performance for a relatively small price adder. Additionally, the top-to-bottom performance gains are significantly more than the Xeon® 5400 series - over 80% for Xeon® 5500 vs 30% for Xeon® 5400. Let’s talk about how these performance gains with high-end SKUs can actually LOWER your TCO.
  • NHM-EP Server cost assumed $6,900 (assumes OEMs charge a premium at launch)  Payback = 21*6900/61,337 = 2year 5month ROI

Transcript

  • 1. Как выбрать оптимальную серверную архитектуру для создания высокоэффективных ЦОД Дмитрий Грязнов Директор по развитию корпоративных проектов в Украине Intel Corporation
  • 2. Отказ от ответственности
    • Intel may make changes to specifications and product descriptions at any time, without notice.
    • Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests.  Any difference in system hardware or software design or configuration may affect actual performance.  Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing.  For more information on performance tests and on the performance of Intel products, visit Intel Performance Benchmark Limitations
    • Intel does not control or audit the design or implementation of third party benchmarks or Web sites referenced in this document. Intel encourages all of its customers to visit the referenced Web sites or others where similar performance benchmarks are reported and confirm whether the referenced benchmarks are accurate and reflect performance of systems available for purchase.
    • Intel processor numbers are not a measure of performance. Processor numbers differentiate features within each processor family, not across different processor families. See www.intel.com/products/processor_number for details.
    • Intel, processors, chipsets, and desktop boards may contain design defects or errors known as errata, which may cause the product to deviate from published specifications. Current characterized errata are available on request.
    • Intel Virtualization Technology requires a computer system with a processor, chipset, BIOS, virtual machine monitor (VMM) and applications enabled for virtualization technology. Functionality, performance or other virtualization technology benefits will vary depending on hardware and software configurations. Virtualization technology-enabled BIOS and VMM applications are currently in development.
    • 64-bit computing on Intel architecture requires a computer system with a processor, chipset, BIOS, operating system, device drivers and applications enabled for Intel® 64 architecture. Performance will vary depending on your hardware and software configurations. Consult with your system vendor for more information.
    • Lead-free: 45nm product is manufactured on a lead-free process.  Lead is below 1000 PPM per EU RoHS directive (2002/95/EC, Annex A).  Some EU RoHS exemptions for lead may apply to other components used in the product package.
    • Halogen-free: Applies only to halogenated flame retardants and PVC in components.  Halogens are below 900 PPM bromine and 900 PPM chlorine.
    • Intel, Intel Xeon, Intel Core microarchitecture, and the Intel logo are trademarks or registered trademarks of Intel Corporation or its subsidiaries in the United States and other countries.
    • © 2009 Standard Performance Evaluation Corporation (SPEC) logo is reprinted with permission
  • 3. Что дает бизнесу обновление ИТ-инфраструктуры?
  • 4. ИНТЕЛЛЕКТУАЛЬНЫЙ ПРОЦЕССОР. НОВАЯ ЭРА ВЫЧИСЛЕНИЙ .
  • 5. Типовой ИТ-бюджет Source: Gartner IT Key Metrics Data 2008. 13% Инновации и новые возможности 20% Раскрытие возможностей установленного оборудования и ПО 67% Текущая эксплуатация Большинство компаний тратит слишком много на поддержку и слишком мало на инновации А как будет потрачен ваш ИТ-бюджет ?
  • 6. « Регулярные инвестиции в новую инфраструктуру , такие как обновление серверов , критически важны . В службе Intel IT, мы установили, что постоянное улучшение существующей у нас инфраструктуры действительно сокращает суммы, необходимые для поддержки наших систем и поэтому является для нас высокоприоритетной инвестицией . » — Diane Bryant, CIO, Intel Corporation
  • 7. Типовой цикл обновления – 3-4 года “ Разработайте программу регулярных изменений для перехода на более энергоэффективное оборудование — в среднем с интервалом в три-пять лет — для управления растущими затратами [ энергии ].” — Gartner, Ноябрь 2007 2 “ Детальный анализ возврата инвестиций (ROI) показал, что мы можем сократить затраты, обновляя все сервера старше четырех лет . Ранее мы использовали сервера значительно дольше с целью максимизации полезной отдачи от них .” — Intel IT Whitepaper, Июнь 2008 1 1 Source: Intel IT Server Refresh whitepaper, June 2008 http://communities.intel.com/docs/DOC-1693 2 Source: Data Center Conference: Day 4 Highlights the ‘Greening’ of Data Centers, by John R. Phelps and Mike Chuba, November 30, 2007, Gartner, Inc., ID Number: G00153658.
  • 8. Преимущества обновления серверов ( одноядерные ЦПУ) Source: Intel estimates as of Nov 2008. Performance comparison using SPECjbb2005 bops (business operations per second). Results have been estimated based on internal Intel analysis and are provided for informational purposes only.  Any difference in system hardware or software design or configuration may affect actual performance. For detailed calculations, configurations and assumptions refer to the legal information slide in backup. 2005 184 Intel ® Xeon ® Single Core Servers 2009 Цель: производительность 1:1 9x рост производительности 184 Intel ® Xeon ® 5500 Based Servers 18% сокращение расходов на электроэнергию в год или Цель: эффективность 9:1 21 Intel ® Xeon ® 5500 Based Servers Окупается за 8 Месяцев 90% сокращение расходов на электроэнергию в год
  • 9. Преимущества обновления серверов ( двухъядерные ЦПУ) 2009 21 Servers 2006 63 Servers 1 Source: Intel internal measurements as of Feb 2009. Performance comparison using SPECjbb2005 bops (business operations per second). Use this slide in conjunction with backup data. 3:1 3:1 5100 Series 5500 Series Intel ® Xeon ® 5500 может снизить затраты на расширение дата-центра, , сократить занимаемое место и другие издержки Требуемое место 66% СОКРАЩЕНИЕ Оценка затрат на энергию 74% СОКРАЩЕНИЕ Оценка за год ЭКОНОМИЯ $61K Энергия + Лицензии ОС
  • 10. Представляем новое семейство ЦПУ Intel ® Xeon ® 5500 и Intel ® Xeon ® 7 500 На базе микроархитектуры Intel ® нового поколения Новое поколение интеллектуальных серверов
  • 11. Процессоры Intel ® Xeon ® серии 5500 ( кодовое имя Nehalem-EP)
    • Невероятный прорыв в технологиях
    • Новая микроархитектура ЦПУ
    • Новая архитектура платформы
    • Новая подсистема памяти
    • Новая подсистема I/O
    • Новая опция – SSD
    • И огромный шаг вперед для бизнеса
    Производительность Энергоэффективность Виртуализация
  • 12. Средства повышения производительности Процессоры intel ® Xeon ® серии 5500
    • Intel ® Turbo Boost Technology
    • Intel ® Hyper-threading Technology
    Core0 Core1 Core2 Core3 Core0 Core1 Core2 Core3 Core0 Core1 Повышает производительность за счет увеличения тактовой частоты ядер при сохранении нормативного TDP Более высокая производительность при необходимости Повышает производительность за счет исполнения одновременного исполнения двух потоков на одном ядре Более высокая производительность для многопоточного ПО Frequency Все ядра работают на номинальной частоте Все ядра работают на более высокой частоте Несколько ядер могут работать на еще более высокой частоте 4C Turbo Normal <4C Turbo
  • 13. Интеллектуальная производительность Новая архитектура серверных платформ Intel ®
    • Интенсивная обработка данных
    • Intel ® QuickPath Technology
    • Integrated Memory Controller
    • Производительность по запросу
    • Intel ® Turbo Boost Technology
    • Intel ® Intelligent Power Technology
    • Многопоточные приложения
    • 45 -нанометровые четырехъядерные ЦПУ Intel ® Xeon ®
    • Intel ® Hyper-threading Technology
    All future products, dates, and figures are preliminary and are subject to change without any notice. Intel ® 5500/ 5520 Chipset PCI Express* 2.0 ICH 9/10 Intel ® X25-E SSDs Intel ® 82599 10GbE Controller Intel ® Data Center Manager Intel ® Node Manager Technology Производительность, соответствующая вашим задачам
  • 14. Производительность процессоров Intel® Xeon™ класса 2 S Single Core Multi-core Relative Performance Intel ® Xeon ® 5100 Intel ® Xeon ® 5300 Intel ® Xeon ® 5400 Intel ® Xeon ® 5500 Performance gains represent a blend (GEO mean) of five common 2-socket workload types across a range of typical usages (see details) Dual Core Source: Published/submitted/measured results March 30, 2009. Each bar represents the geo mean of published results on five industry standard benchmarks – SPECint_rate, SPECfp_rate, SPECjbb2005, TPC-C and SAP-SD. Relative Performance Higher is better Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.
  • 15. Производительность серверов Xeon 5500 в сравнении с Xeon 5400 ERP Floating point Database Web Database Virtualization Server Side Java Java Apps Energy Efficiency Integer App Server Xeon 5400 series Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others. Source: Published/submitted/approved results March 30, 2009. See backup for additional details Relative Performance Higher is better Закрепляет лидерство в серверных тестах
  • 16. Intel ® Xeon ® Processor 7500 Series (Codenamed Nehalem-EX)
    • Nehalem arch optimized for the expandable performance segment
      • New processor architecture
      • New platform architecture
      • New memory subsystem
      • New I/O subsystem
      • New Mission Critical RAS
      • New Levels of Scalability
    • The biggest performance jump ever in Xeon ® history
    Scalable Performance Flexible Virtualization Advanced RAS (Reliability, Availability & Servicability)
  • 17. Intel ® Xeon ® Processor 7500 Series An Intelligent Choice Energy Efficient Refresh Use up to 20:1 server replacement and achieve an estimated < 1 year payback with up to 90% lower operating costs † Performance Refresh Use 1:1 server replacement to deliver over 20x performance gains within an existing data center design † Intel Xeon Processor 7500 Series † Comparisons with single core Intel Xeon based on pre-production measurements of Intel Xeon 7500 compared to single core Xeon from 2005. For notes and disclaimers, see performance and legal information slides at end of this presentation. The Right Investment – Right Now Data Demanding Enterprise, Virtualization or HPC
  • 18. Expanding Platform Capabilities CPU Sockets Max Memory Slots Max I/O Lanes Max Memory Capacity Xeon ® 7400 Xeon ® 7500 * Higher scaling accomplished through use of 3 rd party OEM chipsets. Platform capabilities of node controller systems vary by OEM Xeon ® 7500’s design modularity brings new meaning to &quot;Expandable Platform&quot; 4S >4* 32 OEM dependent 256GB OEM dependent 28 (Gen1) OEM dependent 2 4 8 >8* 32 64 128 OEM dependent 512GB 1TB 2TB OEM dependent 72 (gen2) 72 (gen2) 288 (gen2) OEM dependent
  • 19. Intel “Glueless” & 3 rd Party Node Controller Platforms (examples) Intel “Glueless” (use only Intel chips) Node Controller (use 3 rd Party OEM node controller chips) 4 Socket Intel “glueless” platforms scale up to 8 sockets Higher socket scaling avail via 3 rd Party Node Controllers Node controller configurations are OEM specific 2-socket nodes 4-socket nodes EX EX EX EX EX EX EX EX Node Controller Node Controller OEM specific design Node Controller Node Controller EX EX EX EX EX EX EX EX OEM specific design EX EX EX EX EX EX EX EX 8 Socket
  • 20. Intel ® Xeon ® Processor NEW 6500 Series ^ Sockets supported with Intel platforms only. Higher socket support available via use of 3 rd party OEM node controller * The higher NHM-EX memory capacity can be used to populate for highest capacity/performance or enable use of lower cost lower-density DIMMS. 128 DIMM slots supported on 8 socket systems; 4 sockets systems support 64 DIMMs. New Xeon ® 6500 (2 Socket EX) More processing threads, cache, memory capacity & RAS than Xeon ® 5000 Ideal for highly threaded, big memory databases & virtualization workloads Xeon ® 5500 Xeon ® 5600 Xeon ® 6500 Xeon ® 7500 uArchitecture Nehalem Westmere Nehalem Nehalem Platform Xeon 5500 series Xeon 5500 series Xeon 7500 series Xeon 7500 series Max Sockets Supported 2 2 2^ 8^ Cores/Threads (per socket) 4/8 6/12 8/16 8/16 Cache (level 3) 8 12 18 24 Memory DIMM Slots* 18 18 32 128* RAS Basic Basic Advanced Advanced Target Usage High perf 2S High perf 2S Premium 2S 4S & higher
  • 21. Nehalem-based Server Performance The Greatest Intel® Xeon® Performance Leap In History! Expecting larger gains from Nehalem Architecture in MP 1 Based on May’09 internal measurement using Intel internal workload 2 Based on Sept’09 internal measurement using OLTP workload. NHM-EX vs Dunnington-Caneland. Xeon ® 5500 vs. Xeon ® 5400 Up to 3.5x Memory Bandwidth Up to 2.5x Database Performance Up to 1.7x Integer Throughput Up to 2.2x Floating Point Throughput Nehalem-EX vs. Xeon ® 7400 Up to 9x Memory Bandwidth 1 Up to 3.0x Database Performance 2 > 1.7x Integer Throughput > 2.2x Floating Point Throughput
  • 22. Энергоэффективность новой платформы Ключевые технологии
    • Мониторинг потребления энергии
    • Управление через политики энергосбережения
    Intel ® Intelligent Power Node Manager
    • На уровне группы серверов
    • Агрегированная отчетность
    • Динамическое назначение политик
    Intel ® Dynamic Power Data Center Manager
    • Независимое управление питанием core/uncore
    • Снижение энергопотребления в простое и в среднем
    • Больше режимов энергопотребления
    • Больше режимов сна
    • Управление питанием шины QPI, шины PCIe, и состоянием памяти
    Средства управления энергопотреблением повышают уровень контроля!
  • 23. Линейка процессоров Intel® Xeon™ 5500 † Max Turbo Boost frequency based on number of 133 MHz increments above base freq (+2 = 0.266 GHz, +3 = 0.400 GHz). Дополнительные возможности – в старших моделях
  • 24. Линейка процессоров Intel ® Xeon ® 5500 Производительность Source: Intel internal measurements, January 2009. Relative performance is an average of SPECfp_rate_base2006 and SPECint_rate2006. Subject to change without notice. For notes and disclaimers, see performance and legal information slides at end of this presentation. Performance † Usage Features Top to Bottom Performance † †† Up to +82% Differentiation: Frequency QuickPath CPU Cache TurboBoost Hyper-threading 0.0 1.0 Relative Performance Advanced
    • 6.4 GT/s QPI
    • 8M cache
    • DDR3 1333
    • Turbo +3
    • HT
    X 5560 X5570 Advanced X5550 Standard
    • 5.86 GT/s QPI
    • 8M Cache
    • DDR3 1066
    • Turbo +2
    • HT
    E5520 E5530 E5540 Standard Low Power Option Basic
    • 4.8 GT/s QPI
    • 4M cache
    • DDR3 800
    E5504 E5506 Basic Low Power Option up to 39% up to 13%
  • 25. Особенности выбора памяти
    • 8M cache
    • 6.4 GT/s QPI
    • HT
    • Turbo
    • 8M cache
    • 5.86 GT/s QPI
    • HT
    • Turbo
    • 4M cache
    • 4.8 GT/s QPI
    Advanced DDR3 800 19.2 GB/s 144 GB DDR3 1066 25.5 GB/s 96 GB DDR3 1333 32 GB/s 48 GB Max Capacity Balanced Performance Max Bandwidth HPC General Purpose Enterprise workloads Virtualized Environment E5506 2.13 GHz E5504 2.00 GHz E5502 1.86 GHz (2C) Example Usages Memory Technology E5540 2.53 GHz E5530 2.40 GHz E5520 2.26 GHz X5570 2.93 GHz X5560 2.80 GHz X5550 2.66 GHz E5540 2.53 GHz E5530 2.40 GHz E5520 2.26 GHz X5570 2.93 GHz X5560 2.80 GHz X5550 2.66 GHz Standard Basic X5570 2.93 GHz X5560 2.80 GHz X5550 2.66 GHz 80W 95W 80W Available CPU Options Memory Requirement
  • 26. Заключение † Source: Intel internal measurements, January 2009. For notes and disclaimers, see performance and legal information slides at end of this presentation. Производитель-ность Виртуализация
    • Расширение существующего пула с Intel VT FlexMigration
    • Уникальные возможности для Live Migration
    • Поддержка большего числа VM на одном сервере †
    Энерго-эффективность
    • Рост производительности до 2.25x при сохранении потребления энергии †
    • Сокращение затрат за счет сниженного потребления в простое
    • Динамическое вкл/откл ядер при необходимости
    • Расширение пропускной способности при расчетах до 3.5x †
    • Рост производительности в корпоративных задачах до 2.25x †
    • В 2x больше потоков и автоматический прирост производительности при неободимости
    Intel® Xeon™ 5500 – правильные инвестиции!
  • 27. Процессоры Intel ® Xeon ® серии 5500 Спасибо за внимание! Вопросы? Новое поколение интеллектуальных серверов
  • 28. Linked Foils
  • 29. Legal Information Slides
  • 30. Performance Claim Backup
    • Up to 2.25x performance compared to Xeon 5400 series claim supported by multiple performance results including an OLTP database benchmark and a bandwidth intensive scientific computing benchmark (SPECfp_rate_base2006). Intel internal measurement. (Feb 2009)
      • Configuration details: - OLTP benchmark
      • Baseline platform: Intel preproduction server platform with two Quad-Core Intel® Xeon® processor X5460, 3.16 GHz, 2x6MB L2 cache, 1333MHz system bus, 64GB memory (16x4GB FB DDR2-667), Microsoft Windows Server 2008 Enterprise x64 Edition OS. Performance measured in transactions per second.
      • New platform: Intel preproduction server platform with two Quad-Core Intel® Xeon® processor X5570, 2.93 GHz, 8MB L3 cache, 6.4QPI, 72GB memory (18x4GB DDR3-800), Microsoft Windows Server 2008 Enterprise x64 Edition OS. Performance measured in transactions per second.
      • Configuration details: - SPECfp_rate_base2006 benchmark
      • Baseline platform: Intel preproduction server platform with two Quad-Core Intel® Xeon® processor X5482, 3.20 GHz, 2x6MB L2 cache, 1600MHz system bus, 16GB memory (8x2GB FB DDR2-800), SUSE Linux Enterprise Server 10 SP2 OS. Intel C++ Compiler for Linux32 and Linux64 version 11.0
      • New platform: Intel preproduction server platform with two Quad-Core Intel® Xeon® processor X5570, 2.93 GHz, 8MB L3 cache, 6.4QPI, 24GB memory (6x4GB DDR3-1333), SUSE Linux Enterprise Server 10 SP2 OS. Intel C++ Compiler for Linux32 and Linux64 version 11.0
    • Up to 3.5x platform memory bandwidth compared to Xeon 5400 series claim supported by Stream-Triad results. Intel internal measurement. (Feb 2009)
      • Configuration details: - Stream-Triad benchmark
      • Baseline platform: Intel preproduction server platform with two Quad-Core Intel® Xeon® processor E5472, 3.0 GHz, 2x6MB L2 cache, 1600MHz system bus, 16GB memory (8x2GB FB DDR2-800), Red Hat Enterprise Linux Server 5.3. Stream binaries compiled with Intel compiler 11.0.
      • New platform: Intel preproduction server platform with two Quad-Core Intel® Xeon® processor X5570, 2.93 GHz, 8MB L3 cache, 6.4QPI, 24GB memory (6x4GB DDR3-1333), Red Hat* Enterprise Linux Server 5.3. Stream binaries compiled with Intel compiler 11.0.
  • 31. Eight Month ROI Claim – Back up
    • 8 month ROI claim estimated based on comparison between 2S Single Core Intel® Xeon® 3.80 with 2M L2 Cache and 2S Intel® Xeon® X5570 based servers. Calculation includes analysis based on performance, power, cooling, electricity rates, operating system annual license costs and estimated server costs. This assumes 8kW racks, $0.10 per kWh, cooling costs are 2x the server power consumption costs, operating system license cost of $900/year per server, per server cost of $6900 based on estimated list prices and estimated server utilization rates. All dollar figures are approximate. Performance and power comparisons are based on measured SPECjbb2005* benchmark results (Intel Corporation Feb 2009). Platform power was measured during the steady state window of the benchmark run and at idle. Performance gain compared to baseline was 9x while the platform power was 0.8x.
      • Baseline platform: Intel server platform with two 64-bit Intel Xeon Processor 3.80Ghz with 2M L2 Cache, 800 FSB, 8x1GB DDR2-400 memory, 1 hard drive, 1 power supply, Microsoft* Windows* Server 2003 Ent. SP1, BEA* JRockit* build P27.4.0-windows-x86_64 run with 2 JVM instances
      • New platform: Intel server platform with two quad-core Intel® Xeon® processor X5570, 2.93 GHz, 8MB L3 cache, 6.4QPI, 12 GB memory (6x2GB DDR3-1333), 1 hard drive, 1 power supply, Microsoft Windows Server 2008 Ent. SP1, BEA JRockit build P27.4.0-windows-x86_64 run with 2 JVM instances
    • Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit Intel Performance Benchmark Limitations.
  • 32. Single Core Energy Efficient Refresh Calculation Details For backup, but not required as part of disclaimer 2005 2009 Delta / Notes Product Intel® Xeon® 3.8GHz with 2M cache Intel® Xeon® X5570 (2.93GHz) Performance per Server 1 8.7x increase Intel internal SPECjbb2005* measurements as of Feb 2009 Server Power Idle / Active Power 228W idle / 382W active 174W idle / 312W active Server idle for 16 hours per day and active for 8 hours per day # Servers needed 184 21 ~ 9:1 server consolidation # Racks needed 9 racks 1 rack 9:1 Rack Consolidation Annual Server kWhr 451,474 42,515 90% lower energy costs Total Annual Energy Costs $90,294 $8,502 $81,792 electricity costs per year. Assumes $0.10/kWhr and 2x cooling factor Operating System Licensing Costs $165,600 $18,900 $146,700 less per year Assumes a RHEL 1yr license at $900 Source www.dell.com as of 12/16/08 Annual Cost Savings of $228,854 Cost of new HW n/a $147,000 Assume $6,900 per server Payback Period of 8 months
  • 33. Single Core Performance Refresh Calculation Details 2005 2009 Delta / Notes Product Intel® Xeon® 3.8GHz with 2M cache Intel® Xeon® X5570 (2.93GHz) Performance per Server 1 8.7x increase Intel internal SPECjbb2005* measurements as of Feb 2009 Power Consumption per Server 382W active 312W active Server active 24hr per day Data Center Capability = 1 MW DC Cooling Factor 1.6 PUE 1.6 PUE Same Design PUE = Power Usage Effectiveness. # of Servers 1,636 1,636 Same Footprint Data Center Performance 1 8.7x increase Intel internal SPECjbb2005* measurements as of Feb 2009 Data Center Power 1,000 KW 825 kW 18% Lower Power # of Servers * Svr Power * PUE
  • 34. Benchmark configuration details
    • All comparisons based on published/submitted/approved results as of March 30, 2009
    • SPECint_rate_base2006:
    • Baseline Intel® Xeon® processor X5470 based platform details: Fujitsu Siemens PRIMERGY* RX200 S4 server platform with two Intel Xeon processors X5470 3.33GHz, 12MB L2 cache, 1333MHz FSB, 16GB memory (8x2GB DDR2 PC2-5300F, 2 rank, CAS 5-5-5, with ECC), SUSE Linux Enterprise Server 10 SP2 x86_64 Kernel 2.6.16.60-0.21-smp*, Intel C++ Compiler for Linux32* and Linux64* version 11.0 build 20080730. Referenced as published at 140. For more information see http://www.spec.org/cpu2006/results/res2008q3/cpu2006-20080901-05156.html .
    • Intel® Xeon® processor X5570 based platform details: Fujitsu PRIMERGY* TX300 S5 server platform with two Intel Xeon processors X5570 2.93GHz, 8MB L3 cache, 6.4GT/s QPI, 48 GB memory (6x8 GB PC3-10600R, 2 rank, CL9-9-9, ECC), SUSE Linux Enterprise Server 10 SP2 x86_64 Kernel 2.6.16.60-0.21-smp, Intel C++ Compiler for Linux32 and Linux64 version 11.0 build 20010131. Submitted to www.spec.org for review at 240 as of March 30, 2009.
    • Quad-Core AMD Opteron* processor model 2386SE based platform details: Supermicro A+ Server 1021M-UR+B* Server platform with two Quad-Core AMD Opteron processors 2386SE 2.80GHz, 6MB L3 cache, 32GB memory (8x4GB, DDR2-800, Reg, dual-rank CL5), SUSE Linux Enterprise Server 10 64-bit*, PGI Server Complete Version 7.2 PathScale Compiler Suite*, Release 3.2. Referenced as published at 116. For more information see http://www.spec.org/cpu2006/results/res2009q1/cpu2006-20081218-06302.html .
    • SPECfp_rate_base2006
    • Baseline Intel® Xeon® processor X5482 based platform details: Hewlett-Packard ProLiant DL160 G5p server platform* with two Intel Xeon processors X5482 3.20GHz, 12MB L2 cache, 1600MHz FSB, 16GB memory (8x2GB 800MHz CL5 FB-DIMM), 64-Bit SUSE Linux Enterprise Server 10 SP1, Intel C++ Compiler for Linux32 and Linux64 version 10.1 build 20080730. Referenced as published at 86.4. For more information see http://www.spec.org/cpu2006/results/res2008q4/cpu2006-20081013-05587.html .
    • Intel® Xeon® processor X5570 based platform details: Fujitsu PRIMERGY* TX200 S5 server platform with two Quad-Core Intel Xeon processors X5570 2.93GHz, 8MB L3 cache, 6.4GT/s QPI, 24 GB memory (6x4 GB PC3-10600R, 2 rank, CL9-9-9, ECC), SUSE Linux Enterprise Server 10 SP2 x86_64 Kernel 2.6.16.60-0.21-smp, Intel C++ Compiler for Linux32 and Linux64 version 11.0 build 20010131. Submitted to www.spec.org for review at 194 as of March 30, 2009.
    • Intel® Xeon® processor X5570 based platform details: Cisco B-200 M1 server platform with two Quad-Core Intel Xeon processors X5570 2.93GHz, 8MB L3 cache, 6.4GT/s QPI, 24 GB memory (6x4 GB PC3-10600R, 2 rank, CL9-9-9, ECC), SUSE Linux Enterprise Server 11_RC4 x86_64 2.6.27.15-2-default, Intel C++ Compiler for Linux32 and Linux64 version 11.0 build 20010131. Result measured at 194 as of March 30, 2009.
    • Quad-Core AMD Opteron* processor 2386SE based platform details: SuperMicro A+ Server 1021M-UR+B Server platform* with two Quad-Core AMD Opteron processors 2386SE 2.80GHz, 6MB L3 cache, 32GB memory (8x4GB, DDR2-800, Reg, dual-rank CL5), SUSE Linux Enterprise Server 10 64-bit*, PGI Server* Complete Version 7.2 PathScale Compiler Suite*, Release 3.2. Referenced as published at 107. For more information http://www.spec.org/cpu2006/results/res2009q1/cpu2006-20081218-06303.html .
    • SPECint_base2006:
    • Intel® Xeon® processor X5470 based platform details: Dell PowerEdge 2950 III server platform* with two Intel Xeon processors X5470 3.33GHz, 12MB L2 cache, 1333MHz FSB, 16GB memory (4x4GB DDR2 667MHz FB Dimm), SUSE Linux Enterprise Server 10* SP1 x86_64, kernel 2.6.16.60-0.21-smp*, Intel C++ Compiler for Linux32* and Linux64* version 10.1 build 20080730. Referenced as published at 26.5. For more information see http://www.spec.org/cpu2006/results/res2008q4/cpu2006-20080929-05450.html .
    • Intel® Xeon® processor X5570 based platform details: SuperMicro X8DTN+* server platform with two Intel Xeon processors X5570 2.93GHz, 8MB L3 cache, 6.4GT/s QPI, 24 GB memory (6x4 GB DDR3-1333MHz), SUSE Linux Enterprise Server 10 SP2 x86_64 Kernel 2.6.16.60-0.34-smp, Intel C++ Compiler for Linux32 and Linux64 version 11.0 build 20010131. Submitted to www.spec.org for review at 31.9 as of March 30, 2009.
    • Quad-Core AMD Opteron* processor 2384 based platform details: HP ProLiant* DL385 G5p platform with two AMD Opteron 2384, 6MB L3 cache, 16 GB (4x4 GB, PC2-6400P CL5) memory, SuSE Linux Enterprise Server 10 (x86_64) SP1, Kernel 2.6.16.46-0.12-smp, PGI Server Complete Version 7.2*, PathScale Compiler Suite Version 3.2*. Referenced as published at 17.5. For more information see http://www.spec.org/cpu2006/results/res2008q4/cpu2006-20081110-05930.html .
    Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.
  • 35. Benchmark configuration details
    • All comparisons based on published/submitted/approved results as of March 30, 2009
    • SPECfp_base2006
    • Intel® Xeon® processor X5482 based platform details: IBM System x 3450* server platform with two Intel Xeon processors X5482 3.20GHz, 12MB L2 cache, 1600MHz FSB, 32GB memory, 64-Bit SUSE Linux Enterprise Server 10*, Intel C++ Compiler for Linux32* and Linux64* version 10.1 build 20080730. Referenced as published at 26.5. For more information see http://www.spec.org/cpu2006/results/res2009q1/cpu2006-20081215-06283.html .
    • Intel® Xeon® processor X5570 based platform details: Sun Blade X6270* server platform with two Intel Xeon processors X5570 2.93GHz, 8MB L3 cache, 6.4GT/s QPI; 24GB (6x4 GB DDR3-1333MHz), OpenSolaris* 2008.11, ZFS, Studio 12* Update 1, Auto-parallel tuning. Result submitted to www.spec.org for review at 44.2 as of March 30, 2009.
    • Intel® Xeon® processor X5570 based platform details: SuperMicro X8DTN+* server platform with two Intel Xeon processors X5570 2.93GHz, 8MB L3 cache, 6.4GT/s QPI, 24 GB memory (6x4 GB DDR3-1333MHz), SUSE Linux Enterprise Server 10 SP2 x86_64 Kernel 2.6.16.60-0.34-smp, Intel C++ Compiler for Linux32 and Linux64 version 11.0 build 20010131. Submitted to www.spec.org for review at 38.3 as of March 30, 2009.
    • Quad-Core AMD Opteron* processor model 2384 based platform details: HP ProLiant DL165 G5* platform with two AMD Opteron 2384, 6MB L3 cache, 16 GB (4x4 GB, PC2-6400P CL5) memory, SuSE Linux Enterprise Server 10 (x86_64) SP1, Kernel 2.6.16.46-0.12-smp*, PGI Server Complete Version 7.2*, PathScale Compiler Suite Version 3.2*. Referenced as published at 19.5. For more information see http://www.spec.org/cpu2006/results/res2009q1/cpu2006-20090116-06416.html .
    • SAP-SD 2-Tier
    • Baseline Intel® Xeon® processor X5470 based platform details: HP ProLiant BL460C server platform* with two Intel Xeon processors X5470 3.33GHz, 12MB L2 cache, 1333MHz FSB, 32GB memory, Microsoft Windows Server 2003 Enterprise Edition, Microsoft SQL Server 2005, SAP ECC Release 6.0 . Referenced as published at 2,518 SD users. Certification number 2008048.
    • Intel® Xeon® processor X5570 based platform details: IBM System x3650 M2 Server with two Intel Xeon processors X5570, 2.93GHz 8MB L3 cache, 6.4QPI, 48GB memory, Microsoft Windows Server 2003 Enterprise Edition, DB2 9.5, SAP ECC Release 6.0 (2005). Referenced as published at 5,100 SD users. Certification number 2008079.
    • Quad-Core AMD Opteron* processor model 2384 based platform details: HP Proliant DL385 G5p Server platform* with two Quad-Core AMD Opteron processors 2384 2.70GHz, 6MB L3 cache, 32GB memory, Microsoft Windows Server 2003 Enterprise Edition, Microsoft SQL Server 2005 database, SAP ECC Release 6.0. Referenced as published at 2,752 SD Users. Certification number 2008065.
    • TPC-C
    • Baseline Intel® Xeon® processor X5460 based platform details: HP ProLiant ML370 G5 platform with Intel Xeon processor X5460 3.16GHz (2 processors / 8 cores / 8 threads), 2x6MB L2 cache, 1333 MHz system bus, 64GB memory, Microsoft SQL Server 2005 x64 Enterprise Edition SP2, Microsoft Windows Server 2003 Enterprise x64 Ent. R2. Referenced as published at 275,149 tpmC and $1.44/tpmC; availability date January 7, 2008. For more information see http://tpc.org/tpcc/results/tpcc_result_detail.asp?id=108010701 .
    • Intel® Xeon® processor X5570 based platform details: HP ProLiant DL370 G6* platform with two Intel Xeon processors X5570 2.93GHz (2 processors / 8 cores / 16 threads), 8MB L3 cache, 6.4GT/s QPI, 144 GB memory (18x8 GB DDR3). Oracle 11g database* with Oracle Enterprise Linux OS*. Result submitted to www.tpc.org as of March 30, 2009.
    • TPC-E
    • Baseline Intel® Xeon® processor X5460 based platform details: Fujitsu-Siemens PRIMERGY TX300 S4 server platform* with Intel Xeon processor X5460 3.16GHz (2 processors / 8 cores / 8 threads), 2x6MB L2 cache, 1333 MHz system bus, 64GB memory, Microsoft SQL Server 200 x64 Enterprise Edition*, Microsoft Windows Server 2008 Enterprise x64*. Referenced as published at 317.45 tpsE and $523.49/tpsE; availability date August 30, 2008. For more information see http://www.tpc.org/tpce/results/tpce_result_detail.asp?id=12 .
    • Intel® Xeon® processor X5570 based platform details: Fujitsu-Siemens PRIMERGY RX300 S5 server platform* with two Intel Xeon processors X5570 2.93GHz (2 processors / 8 cores / 16 threads), 8MB L3 cache, 6.4GT/s QPI, 96 GB memory (12x8 GB DDR3-1066), Microsoft SQL Server 2008 x64 Enterprise Edition, Microsoft Windows Server 2008 Enterprise x64. Result submitted to www.tpc.org at 800tpsE and $343.91/tpsE as of March 30, 2009. Availability date April 1, 2009.
    Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.
  • 36. Benchmark configuration details
    • All comparisons based on published/submitted/approved results as of March 30, 2009
    • SPECjbb2005
    • Baseline Intel® Xeon® processor X5470 based platform details: Fujitsu Siemens PRIMERGY RX200 S4 server platform* with two Intel Xeon processors 5470 3.33GHz, 12MB L2 cache, 1333MHz FSB, 16GB memory, Microsoft Windows Server 2008 Enterprise x64 Edition*, Oracle JRockit 6 P28.0.0 (build P28.0.0-8-109238-1.6.0_05-20090130-1408-windows-x86_64) 4 JVM instances. Referenced as published at 368,034 BOPS. For more information see http://www.spec.org/osg/jbb2005/results/res2009q1/jbb2005-20090220-00583.html .
    • Intel® Xeon® processor X5570 based platform details: IBM Bladecenter HS22 Server platform* with two Intel Xeon processors X5570 2.93GHz, 8MB L3 cache, 6.4GT/s QPI, 24 GB memory (6x4 GB DDR-1333MHz), Microsoft Windows Server 2008 Enterprise x64 Edition, IBM J9 2.4 JRE 1.6.0 (build pwa6460sr5-20090323_04(SR5)) run with 4 JVM instances. Result measured at 604417 BOPS - March 30, 2009.
    • Quad-Core AMD Opteron* processor 2386SE based platform details: SuperMicro A+ Server 1021M-UR+B Server platform* with two Quad-Core AMD Opteron processors 2386SE 2.80GHz, 6MB L3 cache, 32GB memory (8x4GB, DDR2-800, Reg, dual-rank CL5), Microsoft Windows Server 2008 Enterprise with SP1 (64-bit), IBM J9 JVM build 2.4, J2RE 1.6.0 IBM J9 2.4. 2 JVM instances. Referenced as published at 383,020 BOPS. For more information see http://www.spec.org/osg/jbb2005/results/res2009q1/jbb2005-20081218-00567.html .
    • SPECweb2005
    • Baseline Intel® Xeon® processor X5460 based platform details: HP Proliant DL380 G5 server platform with two Intel Xeon processors X5460 3.16GHz, 12MB L2 cache, 32GB memory (8x4G 667MHz ECC DDR2 FB-DIMM), RedHat Enterprise Linux 5 (2.6.18-53.el5), Rock Web Server v1.4.6 x86_64. Referenced as published at 29591. For more information see http://www.spec.org/web2005/results/res2008q1/web2005-20080225-00104.html .
    • Intel® Xeon® processor X5570 based platform details: HP ProLiant DL380 G6 platform* with two Intel Xeon processors X5570 2.93GHz, 8MB L3 cache, 6.4GT/s QPI, 144 GB memory (18x8 GB DDR3), Red Hat Enterprise Linux 5.2 .Rock Web Server v1.4.7 (x86_64). Result submitted to www.spec.org for review at 71,045 as of March 30, 2009.
    • Quad-Core AMD Opteron* processor model 2384 based platform details: HP Proliant DL165 G5 Server platform with two Dual-Core AMD Opteron processors 2384 2.70GHz, 6MB L3 cache, memory 32GB (8x4GB, PC2-6400P CL5), RedHat* Enterprise Linux .5.2 (2.6.18-92.el5), Rock Web Server v1.4.7 x86_64. Referenced as published at 39530. For more information see http://www.spec.org/osg/web2005/results/res2009q1/web2005-20090122-00125.html .
    • Fluent: Comparison based on published/submitted results to www.fluent.com/software/fluent/fl6bench/fl6bench_6.4.x/index.htm as of March 30, 2009. All comparisons were using results run on 8 cores within a single machine on dual socket quad-core servers.
    • Baseline Intel® Xeon® processor X5482 based platform details: Supermicro X7DB8+* server platform with two Intel® Xeon® processors X5482 3.20GHz, 12MB L2 cache, 1600MHz FSB, 16GB memory (8x2GB 800MHz DDR2 FB-DIMM), 64-bit RedHat Enterprise Linux 5.3*. Performance measured using Fluent Version 12.0 Beta. (Version 12.0.13)*. Six individual benchmarks are shown as a measure of single node performance. &quot;Overall&quot; performance is the geometric mean of the six individual benchmarks.
    • Intel® Xeon® processor X5570 based platform details: SGI Altix ICE 8200EX* server platform with two Intel Xeon processors X5570 2.93GHz, 8MB L3 cache, QPI 6.4 MT/sec, 24GB memory (12x2GB 1066MHz DDR3), 64-bit Suse Linux Enterprise Server* 10 SP2 with ProPack 6SP2*. Performance measured using Fluent Version 12.0 Beta. (Version 12.0.9) Six individual benchmarks are shown as a measure of single node performance. &quot;Overall&quot; performance is the geometric mean of the six individual benchmarks.
    • Quad-Core AMD Opteron* processor model 2384 platform based details:Server platform with two AMD Opteron 2384 processor 2.7GHz, 6MB L3 cache, Linux OS. Performance measured using Fluent Version 12.0 Beta. (Version 12.0.7) Six individual benchmarks are shown as a measure of single node performance. &quot;Overall&quot; performance is the geometric mean of the six individual benchmarks.
    Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.
  • 37. Benchmark configuration details
    • All comparisons based on published/submitted/approved results as of March 30, 2009
    • SPECompM2001
    • Baseline Intel® Xeon® processor E5472 based platform details: Supermicro X7DB8+ server platform* with two Intel Xeon processors E5472 3.0GHz, 12MB L2 cache, 1600MHz FSB, 32GB memory (8x4GB 800MHz DDR2 FB-DIMM), SUSE LINUX 10.1* (X86-64) (Linux 2.6.16.13-4-smp). Binaries built with Intel Compiler 10.1. Referenced as published at 17187. (SPECompMbase2001). For more information see http://www.spec.org/omp/results/res2007q4/omp2001-20071107-00274.html .
    • Intel® Xeon® processor X5570 based platform details: Cisco B-200 M1 server platform* with two Intel Xeon processors X5570 2.93GHz, 8MB L3 cache, 6.4GT/s QPI, 24 GB memory (6x4 GB DDR3-1333MHz), Red Hat EL 5.3, Linux Kernel 2.6.18-128.el5 SMP x86_64, Binaries built with Intel® C/C++ Compiler 11.0 for Linux. Result submitted to www.spec.org for review at 43593 (SPECompMbase2001) as of March 30, 2009.
    • Quad-Core AMD Opteron processor 2384 based platform* details: Supermicro H8DMU Server platform* with two Quad-Core AMD Opteron processors 2386SE* 2.80GHz, 6MB L3 cache, 16GB memory (8x2GB, PC2-6400, Reg, dual-rank CL5), SUSE Linux Enterprise Server 10 64-bit, Binaries built with PathScale Compiler Suite*, Release 3.1. Referenced as published at 22678 (SPECompMbase2001). For more information http://www.spec.org/omp/results/res2008q4/omp2001-20081021-00320.html .
    • SPECpower_ssj2008
    • Baseline Intel® Xeon® processor L5430 based platform details: Powerleader PR2510D2 server platform* with two Intel Xeon processors L5430 2.33GHz, 12MB L2 cache, 1333MHz FSB, 8GB memory, Oracle JRockit* (build P27.5.0-5_o_CR371811_CR374296-100684-1.6.0_03-20080702-1651-windows-x86_64, compiled mode). Published at 1135 ssj_ops/watt. For more information see: http://www.spec.org/power_ssj2008/results/res2008q4/power_ssj2008-20081007-00086.html .
    • Intel® Xeon® processor X5570 based platform details: Verari Systems, Inc. VB1305 server platform* with two Intel Xeon processor X5570, 2.93GHz, 8 GB (4 x 2), Microsoft Windows Server 2008 Enterprise* Service Pack 2 OS. Oracle JVM (build P28.0.0-14-111048-1.6.0_05-20090303-1104-windows-x86_64, (compiled mode) result of 1943 provided by Verari as of 3/30/2009.
    • Quad-Core AMD Opteron* processor model 2376HE based platform based details: Supermicro Inc. 1021M-UR+B Server platform* with two Quad-Core AMD Opteron processors 2376HE 2.30GHz, 6MB L3 cache, 16GB memory, IBM J9 VM (build 2.4, J2RE 1.6.0 IBM J9 2.4 Linux amd64-64 jvmxa6460-20081105_25433 (JIT enabled, AOT enabled), Published at 1044 ssj_ops/watt. For more information see: http://www.spec.org/power_ssj2008/results/res2009q1/power_ssj2008-20090210-00111.html .
    • SPECjAppServer2004
    • Baseline Intel® Xeon® processor X5460 based platform details: HP Proliant BL460c G1 server platform with two Intel Xeon processors X5460 3.16GHz, 12MB L2 cache, 16GB memory (8x2G 667MHz ECC DDR2 FB-DIMM), Oracle Application Server 10G Release 10.1.3.3 - Java Edition, BEA JRockit(R) 6.0 JDK (R27.3.0-106) (Linux x86 32bit), Oracle Database Enterprise Edition Release 11.1.0.6. Referenced as published at 2056. For more information see http://www.spec.org/osg/jAppServer2004/results/res2007q4/jAppServer2004-20071023-00088.html .
    • Intel® Xeon® processor X5570 based platform details: Dell PowerEdge R610 server platform* with two Intel Xeon processors X5570 2.93GHz, 8MB L3 cache, 6.4GT/s QPI, 24 GB memory (12x2 GB DDR3), Oracle WebLogic Server Standard Edition Release 10.3, Oracle JRockit(R) 6.0 JDK (R27.6.0-50) (Linux x86 32bit), Oracle Database Enterprise Edition Release 11.1.0.7, Result submitted to www.spec.org for review at 3975 as of March 30, 2009.
    • VMmark:
    • Baseline Intel® Xeon® processor X5470 based platform details: HP Proliant* ML370 G5 server platform with two Intel Xeon processors X5470 3.33GHz, 2x6MB L2 cache, 1333MHz FSB, 48GB memory, VMware ESX V3.5. Update 3 Published at 9.15@7 tiles. For more information see www.vmware.com/files/pdf/vmmark/VMmark-HP-2008-10-09-ML370.pdf .
    • Intel® Xeon® processor X5570 based platform details: Dell PowerEdge* R710 Server platform with two Intel Xeon processors X5570 2.93GHz, 8MB L3 cache, 6.4GT/s QPI, 96 GB memory (12x8 GB DDR3-1066MHz), VMware ESX beta build 150817. Submitted to VMware for review at 23.55@16 tiles.
    • AMD Opteron* processor model 2384 based platform details: HP ProLiant* DL385 G5p platform with two AMD Opteron 2384, 6MB L3 Cache, 16 GB 32GB memory. VMware ESX V3.5. Update 3 Published at 11.28@8 tiles. For more information see www.vmware.com/files/pdf/vmmark/VMmark-HP-2008-12-30-DL385.pdf
    Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.
  • 38. Benchmark configuration details
    • All comparisons based on published/submitted/approved results as of March 30, 2009
    • Reservoir Simulation using Schlumberger Eclipse*
    • Intel® Xeon® processor X5482 based platform details: Supermicro X7DB8+ server platform* with two Intel Xeon processors X5482 3.20GHz, 12MB L2 cache, 1600MHz FSB, 16GB memory (8x2GB 800MHz DDR2 FB-DIMM), 64-bit RedHat Enterprise Linux 5.3*. (64-bit) and Eclipse version 2008.1 software.
    • Intel® Xeon® processor X5570 based platform details: Supermicro X8DTN+ server platform* with two Intel Xeon processors X5570 2.93GHz, 8MB L3 cache, QPI 6.4* MT/sec, 24GB (12x2GB 1066MHz DDR3) memory, 64-bit RedHat Enterprise Linux 5.3. Eclipse version 2008.1 software.
    • Reservoir Simulation using Landmark Nexus*
    • Intel® Xeon® processor X5482 based platform details: Supermicro X7DB8+ server platform* with two Intel Xeon processors X5482 3.20GHz, 12MB L2 cache, 1600MHz FSB, 16GB memory (8x2GB 800MHz DDR2 FB-DIMM), 64-bit RedHat Enterprise Linux 5.2*. (64-bit) . Landmark Nexus R5000 software*.
    • Intel® Xeon® processor X5560 based platform details: Supermicro X8DTN+ server platform* with two Intel Xeon processors X5560 2.80GHz, 8MB L3 cache, QPI 6.4 MT/sec, 12GB memory, 64-bit RedHat Enterprise Linux 5.3. Landmark Nexus R5000 software.
    • Reservoir Simulation using CMG* IMEX*
    • Intel® Xeon® processor X5482 based platform details: Dell Precision T7400 platform* with two Intel Xeon processors X5482 3.20GHz, 12MB L2 cache, 1600MHz FSB, 32GB RedHat Enterprise Linux 5.2*. (64-bit) CMG IMEX, Version 2008.11.
    • Intel® Xeon® processor X5570 based platform details: Supermicro X8DTN+ server platform* with two Intel Xeon processors X5570 2.93GHz, 8MB L3 cache, QPI 6.4* MT/sec, 18GB memory. RedHat Enterprise Linux 5.3. (64-bit) CMG IMEX, Version 2008.11.
    • Computational Fluid Dynamics analysis using Star-CD* (Single Node) - Comparison based on published/submitted results to http://www.cd-adapco.com/products/STAR-CD/performance/406/index.html as of March 30, 2009. All comparisons were using results run on 8 cores within a single machine on dual socket quad-core servers.
    • Intel® Xeon® processor X5482 based platform details: Supermicro X7DB8+* server platform with two Intel Xeon processors X5482 3.20GHz, 12MB L2 cache, 1600MHz FSB, 16GB memory (8x2GB 800MHz DDR2 FB-DIMM), 64-bit RedHat Enterprise Linux 5.3*. Performance measured using STAR-CD v4.06. Same configuration used for all both benchmark results - A-Class and C-Class.
    • Intel® Xeon® processor X5570 based platform details: Supermicro X8DTN+* server platform with two Intel Xeon processors X5570 2.93GHz, 8MB L3 cache, QPI* 6.4 MT/sec, 24GB memory (12x2GB 1066MHz DDR3), 64-bit RedHat Enterprise Linux 5.3. Performance measured using STAR-CD v4.06. Same configuration used for all both benchmark results - A-Class and C-Class.
    • Crash Simulation analysis using LS-DYNA* (Single Node): Comparison based on published/submitted results to http://www.topcrunch.org/ as of March 30, 2009. All comparisons were using results run on 8 cores within a single machine on dual socket quad-core servers.
    • Intel® Xeon® processor X5482 based platform details: Supermicro X7DB8+* server platform with two Intel® Xeon® processors X5482 3.20GHz, 12MB L2 cache, 1600MHz FSB, 16GB memory (8x2GB 800MHz DDR2 FB-DIMM), 64-bit RedHat Enterprise Linux* 5.3. Performance measured using LS-DYNA mpp971.s.R321. Same configuration used for all three benchmark results - neon_refined_revised, 3 vehicle collision, car2car.
    • Intel® Xeon® processor X5570 based platform details: Supermicro X8DTN+ server platform with two Intel Xeon processors X5570 2.93GHz, 8MB L3 cache, QPI 6.4 MT/sec, 24GB memory (12x2GB 1066MHz DDR3), 64-bit RedHat Enterprise Linux 5.3. Performance measured using LS-DYNA mpp971.s.R321. Same configuration used for all three benchmark results - neon_refined_revised, 3 vehicle collision, car2car.
    • SPECompL2001
    • Baseline Intel® Xeon® processor E5472 based platform details: Supermicro X7DB8+ server platform* with two Intel Xeon processors X5482 3.20GHz, 12MB L2 cache, 1600MHz FSB, 32GB memory (8x4GB 800MHz DDR2 FB-DIMM), SUSE LINUX 10.1* (X86-64) Intel Compiler 11.0. Submitted to www.spec.org for review at 81332 as of March 30, 2009.
    • Intel® Xeon® processor X5570 based platform details: Cisco B-200 M1 server platform* with two Intel Xeon processors X5570 2.93GHz, 8MB L3 cache, 6.4GT/s QPI, 24 GB memory (6x4 GB DDR3-1333MHz), Red Hat EL 5.3, Linux Kernel 2.6.18-128.el5 SMP x86_64, Binaries built with Intel® C/C++ Compiler 11.0 for Linux. Result submitted to www.spec.org for review at 234,996 (SPECompMbase2001) as of March 30, 2009.
    Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.
  • 39. Benchmark configuration details
    • All comparisons based on published/submitted/approved results as of March 30, 2009
    • Multiphysics Finite Element Analysis using ANSYS* - Comparison based on published/submitted results to www.ansys.com/services/hardware-support-db.htm as of March 30, 2009.
    • Intel® Xeon® processor X5482 based platform details: Supermicro X7DB8+* server platform with two Intel® Xeon® processors X5482 3.20GHz, 12MB L2 cache, 1600MHz FSB, 16GB memory (8x2GB 800MHz DDR2 FB-DIMM), 64-bit RedHat Enterprise Linux 5.1*. Performance measured using ANSYS* Mechanical* 12.0 Preview 7. Benchmark for Ansys-Shared* consists of a suite of 8 workloads and Ansys-Distributed* consists of a suite of 7 workloads. Geo mean of each these workload groups used for comparison.
    • Intel® Xeon® processor X5570 based platform details: Supermicro X8DTN+ server platform with two Intel Xeon processors X5570 2.93GHz, 8MB L3 cache, QPI 6.4 MT/sec, 24GB memory (12x2GB 1066MHz DDR3), 64-bit RedHat Enterprise Linux 5.3. Performance measured using ANSYS* Mechanical* 12.0 Preview 7. Benchmark for Ansys-Shared consists of a suite of 8 workloads and Ansys-Distributed consists of a suite of 7 workloads. Geo mean of each these workload groups used for comparison.
    • MM5 v4.7.4 - t3a and WRF v3.0.1 - 12km CONUS : Comparison based on measured results as of March 30, 2009. All comparisons were using results run on 8 cores within a single machine on dual socket quad-core servers. Same platform used for both benchmarks
    • Baseline Intel® Xeon® processor X5482 based platform details: SGI Altix ICE 8200EX* server platform with two Intel® Xeon® processors X5482 3.20GHz, 12MB L2 cache, 1600MHz FSB, 16GB memory (8x2GB 800MHz DDR2 FB-DIMM), 64-bit Suse Linux Enterprise Server* 10 SP2 with ProPack 6SP2*.
    • Intel® Xeon® processor X5570 based platform details: SGI Altix ICE 8200EX* server platform with two Intel Xeon processors X5570 2.93GHz, 8MB L3 cache, QPI 6.4 MT/sec, 24GB memory (12x2GB 1066MHz DDR3), 64-bit Suse Linux Enterprise Server* 10 SP2 with ProPack 6SP2*..
    • Unified Finite Element Analysis using Abaqus*: Comparison based on published/submitted results to www.simulia.com/support/v68/v68_performance.php as of March 30, 2009. Geo mean of S2A, S4B, S4D, S6 used for comparison
    • Intel® Xeon® processor X5482 based platform details: Supermicro X7DB8+* server platform with two Intel® Xeon® processors X5482 3.20GHz, 12MB L2 cache, 1600MHz FSB, 16GB memory (8x2GB 800MHz DDR2 FB-DIMM), 64-bit RedHat Enterprise Linux 5.3*. Performance measured using Abaqus/Standard* and Abaqus/Explicit v6.8*-EF1.
    • Intel® Xeon® processor X5570 based platform details: Supermicro X8DTN+ server platform with two Intel Xeon processors X5570 2.93GHz, 8MB L3 cache, QPI 6.4 MT/sec, 24GB memory (12x2GB 1066MHz DDR3), 64-bit RedHat Enterprise Linux 5.3. Performance measured using Abaqus/Standard and Abaqus/Explicit v6.8-EF1.
    • Quad-Core AMD Opteron* processor 2384 based platform details: HP ProLiant* BL465 G5 platform with two AMD Opteron 2384, 6MB L3 cache, 16 GB memory, SuSE Linux Enterprise Server 10 (x86_64).
    • SPECjvm2008:
    • Intel® Xeon® processor X5470 based platform details: Supermicro X7DBE +* server platform with two Intel® Xeon® processors X5470 3.333GHz, 12MB L2 Cache, 1333MHz FSB, 16GB memory (8x2GB 667MHz DDR2 FB-DIMM), Red Hat Enterprise Linux 5.3 (kernel: 2.6.18-128.el5), JDK 1.6.0_06-p-b04. Intel internal measurement. Measured at 179.53 as of March 30, 2009
    • Intel® Xeon® processor X5570 based platform details: Sun Blade X6270* server platform with two Intel Xeon processors X5570 2.93GHz, 8MB L3 cache, 6.4GT/s QPI; 24GB (6x4 GB DDR3-1333MHz), OpenSolaris* 2008.11, ZFS, JDK 1.6.0_14. Result submitted to www.spec.org for review at 296.77 SPECjvm2008 base as of March 30, 2009.
    Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit http://www.intel.com/performance/resources/limits.htm Copyright © 2009, Intel Corporation. * Other names and brands may be claimed as the property of others.
  • 40. “ Alternative to RISC” Legal Disclaimers
    • IBM 4.7 GHz Power6* 2S:
    • SPECint_rate2006*: 2 4.7 GHz Power6* processors p570 server; 2 chips, 2 cores/chip, 2 threads/core; 32 GB DDR2 667 MHz memory, IBM AIX 5L v5.3, XL C/C++ Enterprise Edition V9.0 for AIX compiler; SPECint_rate_base2006* 106, SPECint_rate2006* 122; http://www.spec.org/cpu2006/results/res2007q2/cpu2006-20070518-01100.html
    • SPECfp_rate2006*: 2 4.7 GHz Power6* processors p570 server; 2 chips, 2 cores/chip, 2 threads/core; 32 GB DDR2 667 MHz memory, IBM AIX 5L v5.3, XL C/C++ Enterprise Edition V9.0 for AIX compiler; SPECfp_rate_base2006* 102, SPECfp_rate2006* 115; http://www.spec.org/cpu2006/results/res2007q2/cpu2006-20070518-01101.html
    • SPECjbb2005*: 2 4.7 GHz Power6* processors p570 server; 2 chips, 2 cores/chip, 2 threads/core; 32 GB DDR2 667 MHz memory, IBM AIX 5L v5.3; J2RE 1.6.0 (32-bit) IBM J9 2.4 AIX [build pap3260sr1-20080117_01]; SPECjbb2005* bops=205917, SPECjbb2005* bops/JVM=102959; http://www.spec.org/jbb2005/results/res2008q2/jbb2005-20080604-00502.html
    • SAP SD 2-Tier: 2 4.7 GHz Power Power6* processors p570 server; 2 chips, 2 cores/chip, 2 threads/core; 32 GB DDR2 667 MHz memory, IBM AIX 5L v5.3,
    • Number of benchmark users & comp.: 2,035 SD (Sales & Distribution), Average dialog response time: 1.99 seconds, Throughput: Fully Processed Order Line items/hour: 203,670, Dialog steps/hour: 611,000, SAPS: 10,180, Average DB request time (dia/upd): 0.011 sec / 0.015 sec, CPU utilization of central server: 99%, RDBMS: Oracle 10g, SAP ECC Release: 6.0; http://download.sap.com/download.epd?context=40E2D9D5E00EEF7CE70C8803CE675683DDB114F2D2EBD812EB10A8068EB35C8A
  • 41. (2009) 2 Year, 5 Month ROI Claim – Back up
    • 2 year, 5 month (29 month) ROI claim estimated based on comparison between 2S Dual Core Intel® Xeon® 5160 (3.0GHz) and 2S Intel® Xeon® X5570 based servers. Calculation includes analysis based on performance, power, cooling, electricity rates, operating system annual license costs and estimated server costs. This assumes 8kW racks, $0.10 per kWh, cooling costs are 2x the server power consumption costs, operating system license cost of $900/year per server, per server cost of $6900 based on estimated list prices and estimated server utilization rates. All dollar figures are approximate. Performance and power comparisons are based on measured SPECjbb2005* benchmark results (Intel Corporation Feb 2009). Platform power was measured during the steady state window of the benchmark run and at idle. Performance gain compared to baseline was 9x while the platform power was 0.8x.
      • Baseline platform: Intel server platform with two dual core Intel® Xeon® Processor 5160, 2.93GHz, 1333MHz FSB, 8x2GB FBDMIMM DDR2-667 memory, 1 hard drive, 1 power supply, Microsoft* Windows* Server 2003 Ent. SP1, BEA* JRockit* build P27.4.0-windows-x86_64 run with 2 JVM instances
      • New platform: Intel server platform with two quad-core Intel® Xeon® processor X5570, 2.93 GHz, 8MB L3 cache, 6.4QPI, 24GB memory (6x2GB DDR3-1333), 1 hard drive, 1 power supply, Microsoft Windows Server 2008 Ent. SP1, BEA JRockit build P27.4.0-windows-x86_64 run with 2 JVM instances
    • Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, visit Intel Performance Benchmark Limitations.
  • 42. Dual Core Energy Efficient Refresh Calculation Details For backup, but not required as part of disclaimer 2005 2009 Delta / Notes Product Intel Xeon 5100 series (3.00GHz) Intel Xeon 5500 series (2.93GHz) Performance per Server 1 3x increase Intel internal SPECjbb2005* measurements as of Feb 2009 Server Power Idle / Active Power 252W idle / 354W active 151W idle / 312W active Server idle for 16 hours per day and active for 8 hours per day # Servers needed 63 21 3:1 server consolidation # Racks needed 6 racks 1 rack 3:1 Rack Consolidation Annual kWhr 158,270 40,582 74% lower energy costs Annual Energy Costs $31,654 $8,116 $23,537 electricity costs per year. Assumes $0.10/kWhr and 2x cooling factor OS Licensing Costs $56,700 $18,900 $37,800 less per year Assumes a RHEL 1yr license at $900 Source www.dell.com as of 12/16/08 Annual Cost Savings of $ Cost of new HW n/a $144,900 Payback Period of 2yrs 5months (< 3 years)