POWER8 Performance Analysis

Total Page:16

File Type:pdf, Size:1020Kb

POWER8 Performance Analysis POWER8 Performance Analysis Satish Kumar Sadasivam Senior Performance Engineer, Master Inventor IBM Systems and Technology Labs [email protected] #OpenPOWERSummit Join the conversation at #OpenPOWERSummit 1 Overview . POWER8 Overview . Introduction to Performance Monitoring . Performance Monitoring Features in POWER8 . What’s new in POWER8? . POWER8 Pipeline . CPI Stack overview – Stall Accounting Model . Performance analysis . CPI analysis . Data source analysis . Prefetch control & Prefetch effectiveness . Application level performance analysis . Marked event profiling & performance analysis. Microarchitecture bottleneck analysis . Core bottleneck analysis using trace tool and scroll pipe. Join the conversation at #OpenPOWERSummit 2 POWER8 Processor Join the conversation at #OpenPOWERSummit 3 Improvements over POWER7 Join the conversation at #OpenPOWERSummit 4 Cache Improvements Join the conversation at #OpenPOWERSummit 5 Cache Bandwidths Join the conversation at #OpenPOWERSummit 6 Memory Organization Join the conversation at #OpenPOWERSummit 7 Performance Instrumentation in P8 • Hardware Performance Monitoring is critical to enable performance evaluation of applications/programs on complex performance cores such as POWER8 • POWER8 provides advanced instrumentation capabilities in two layers • Core Instrumentation • Nest level Instrumentation Core Level Nest Level Performance Performance Monitoring Monitoring Join the conversation at #OpenPOWERSummit 8 Core Level Performance Monitoring . Key to root cause performance bottlenecks at core or thread level . Facilitates monitoring of • Core Pipeline efficiency – frontend, branch prediction, execution units, schedulers, etc • Behavior metrics – stalls, execution rates, utilizations, thread prioritization & resource sharing . Enables understanding and optimization of application performance at processor and compiler level. Join the conversation at #OpenPOWERSummit 9 Nest Level Instrumentation . Instrumentation at • L3 Cache, • Interconnect Fabric • Memory channels/controller . Information provided at per-core and chip-level( as against thread-level for core-level counters) . Significance & Usefulness: • Bandwidth Analysis • Key for analyzing the Cloud Virtualized environment performance. • Can be used to effectively monitor the memory and chip level characteristics to employ effective provisioning of the cloud space. Join the conversation at #OpenPOWERSummit 10 What’s new in POWER8? . Enhanced CPI Stack Cycle Accounting Model . Hotness Table . Branch History Rolling Buffer . Event-Based Branches . Prefetch effectiveness events . Additional Events to capture & analyze hardware level performance issues Join the conversation at #OpenPOWERSummit 11 POWER8 Microarchitecture Join the conversation at #OpenPOWERSummit 12 POWER8 Core Pipeline Front end stalls: cycles a thread’s GCT was empty , i.e. pipeline was empty for that thread. Back end stalls: cycles thread had GCT entries but no completion occurred. Join the conversation at #OpenPOWERSummit 13 POWER8 Group Formation . Group formation: • Instructions are formed into groups for dispatch and completion tracking after Instruction Fetch. • Thread priority logic selects up to 8 instructions from the Instruction buffers for group formation in each cycle • Group formation driven by group formation rules . Global Completion Table(GCT) . Completion based performance bottleneck analysis Join the conversation at #OpenPOWERSummit 14 CPI Analysis . Cycles-per-instruction(CPI) stack presents a picture of a typical instruction’s lifespan from fetch to completion . Provides information to narrow down to the bottleneck point(s) in the processor pipeline . POWER8 features a Completion-based CPI Stack accounting model . Time spent in the execution is split into : . Group Completion cycles . Stall cycles Join the conversation at #OpenPOWERSummit 15 POWER8 CPI Stack Stall due to Branch Stall due to BR or CR Stall due to CR Stall due to Fixed-Point Long Stall due to Fixed-Point Stall due to Fixed-Point (Other) Stall due to Vector Long Stall due to Vector Stall due to Vector (other) Stall due to Vector/Scalar Stall due to Scalar Long Stall due to Scalar Stall due to Scalar (other) Completion Stalls Stall due to Vector/Scalar (other) Stall due to Dcache Miss Stall due to LSU Reject Stall due to Store Finish Stall due to LSU Stall due to Load Finish Stall due to Store Forward Stall due to Load/Store (other) Stall due to Next-to-Complete Flush Cycles Waiting to Complete Blocked due to LWSync Blocked due to HWSync Blocked due to ECC Delay Thread Blocked Blocked due to Flush Blocked due to COQ Full Thread Blocked (other) Completion Table Empty due to Completion Table Empty due to IC L3 Miss IC Miss Completion Table Empty due to IC Miss (other) Completion Table Empty due to Branch Mispredict Completion Table Empty due to Branch Mispredict + IC Miss Completion Table Empty Dispatch Held due to Mapper Completion Table Empty – Dispatch Held due to Store Queue Dispatch Held Dispatch Held due to Issue Queue Dispatch Held (other) Completion Table Empty (Other) Completion Cycles Join the conversation at #OpenPOWERSummit CPI Stack – LSU Stalls Join the conversation at #OpenPOWERSummit 17 An Example of CPI Stack CPI Stack 3.000 2.500 PM_CMPLU_STALL 2.000 PM_NTCG_ALL_FIN 1.500 PM_CMPLU_STALL_THRD PM_GCT_NOSLOT_CYC 1.000 PM_GRP_CMPL 0.500 0.000 Prefetch OFF Prefetch ON Join the conversation at #OpenPOWERSummit 18 CPI Stack – Detailed Stall Distribution Completion Stall Components 4.000 PM_CMPLU_STALL_BRU_CRU 3.500 PM_CMPLU_STALL_FXU 3.000 PM_CMPLU_STALL_VSU PM_CMPLU_STALL_VECTOR 2.500 PM_CMPLU_STALL_SCALAR 2.000 PM_CMPLU_STALL_NTCG_FLUSH PM_CMPLU_STALL_LSU 1.500 PM_CMPLU_STALL_DCACHE_MISS 1.000 PM_CMPLU_STALL_REJECT PM_CMPLU_STALL_STORE 0.500 PM_CMPLU_STALL_LOAD_FINISH 0.000 PM_CMPLU_STALL_ST_FWD Prefetch OFF Prefetch ON Join the conversation at #OpenPOWERSummit 19 Data Source Analysis . Analysis of application data accesses across the Cache & Memory hierarchy is key to understanding the following • Performance limiting factors & resource requirements of the application • Scaling capabilities(in multi-threaded scenarios) . Cache hierarchy latencies: Join the conversation at #OpenPOWERSummit 20 Prefetch Controls . Prefetch effects: • Positive . Brings data closer to the core . Reduces memory access stalls • Possible negative effects: . Extra Bandwidth consumption - choking other application memory accesses . Cache pollution . Increased power consumption . POWER8 supports L1 and L3 levels Prefetches . DSCR Register ( Power ISA v2.07 ) DPFD: Default Prefetch Depth SSE: Store Stream Enable SNSE: Stride-N Stream Enable LSD: Load Stream Disable URG: Depth Attainment Urgency Join the conversation at #OpenPOWERSummit 21 Studying Prefetch Effectiveness . POWER8 provides performance events to study the prefetch effectiveness . Counters indicate usage and non-usage of cache lines that are prefetched into the cache at the time of eviction from the cache . Counters available: • MEPF Metrics are used to evaluate the Prefetch effectiveness in POWER8 Join the conversation at #OpenPOWERSummit 22 Application Profiling tools . Market Event Profiling: • Pinpoint performance inhibiting behavior/bottlenecks to specific instruction in application code . Why necessary? • Non-marked events are best suited to study performance metrics • In an OOO super-scalar multiple-issue processor, the profile data from non-marked events can only indicate code “region” responsible for performance bottlenecks • Code “region” granularity can range from few to tens of instructions. Join the conversation at #OpenPOWERSummit 23 Example of Marked Event profiling Join the conversation at #OpenPOWERSummit 24 Marked Events – a non-exhaustive list . PM_MRK_LD_MISS_L1 . PM_MRK_LD_MISS_L1_CYC . PM_MRK_BR_MPRED_CMPL . PM_MRK_BR_TAKEN_CMPL . PM_MRK_DATA_FROM_MEM . PM_MRK_LSU_REJECT . PM_MRK_STCX_FAIL . PM_MRK_GRP_IC_MISS . PM_MRK_DTLB_MISS . PM_MRK_ST_FWD . PM_MRK_LSU_FLUSH . PM_MRK_LSU_FLUSH_ULD . PM_MRK_LSU_FLUSH_UST Join the conversation at #OpenPOWERSummit 25 Microarchitecture Analysis . Deep-dive analysis to root-cause performance inhibitor at processor pipeline stages. Tools used: • Itrace • Cycle Accurate Simulator Trace application with valgrind Generate qtrace Microarchitecture Stats Analyze & simppc Optimize Scrollpipe Application code Join the conversation at #OpenPOWERSummit 26 Tools for Microarchitecture Analysis . IBM SDK for Linux on Power . IBM POWER8 Functional Simulator (systemsim) . Valgrind framework provides application/program tracing capabilities (itrace) . POWER8 Performance Simulator (sim_ppc) https://www-304.ibm.com/webapp/set2/sas/f/lopdiags/sdklop.html Join the conversation at #OpenPOWERSummit 27 Thank You! Join the conversation at #OpenPOWERSummit 28 .
Recommended publications
  • IBM Power System POWER8 Facts and Features
    IBM Power Systems IBM Power System POWER8 Facts and Features April 29, 2014 IBM Power Systems™ servers and IBM BladeCenter® blade servers using IBM POWER7® and POWER7+® processors are described in a separate Facts and Features report dated July 2013 (POB03022-USEN-28). IBM Power Systems™ servers and IBM BladeCenter® blade servers using IBM POWER6® and POWER6+™ processors are described in a separate Facts and Features report dated April 2010 (POB03004-USEN-14). 1 IBM Power Systems Table of Contents IBM Power System S812L 4 IBM Power System S822 and IBM Power System S822L 5 IBM Power System S814 and IBM Power System S824 6 System Unit Details 7 Server I/O Drawers & Attachment 8 Physical Planning Characteristics 9 Warranty / Installation 10 Power Systems Software Support 11 Performance Notes & More Information 12 These notes apply to the description tables for the pages which follow: Y Standard / Supported Optional Optionally Available / Supported N/A or - Not Available / Supported or Not Applicable SOD Statement of General Direction announced SLES SUSE Linux Enterprise Server RHEL Red Hat Enterprise Linux a One x8 PCIe slots must contain a 4-port 1Gb Ethernet LAN available for client use b Use of expanded function storage backplane uses one PCIe slot Backplane provides dual high performance SAS controllers with 1.8 GB write cache expanded up to 7.2 GB with c compression plus Easy Tier function plus two SAS ports for running an EXP24S drawer d Full benchmark results are located at ibm.com/systems/power/hardware/reports/system_perf.html e Option is supported on IBM i only through VIOS.
    [Show full text]
  • Openpower AI CERN V1.Pdf
    Moore’s Law Processor Technology Firmware / OS Linux Accelerator sSoftware OpenStack Storage Network ... Price/Performance POWER8 2000 2020 DRAM Memory Chips Buffer Power8: Up to 12 Cores, up to 96 Threads L1, L2, L3 + L4 Caches Up to 1 TB per socket https://www.ibm.com/blogs/syst Up to 230 GB/s sustained memory ems/power-systems- openpower-enable- bandwidth acceleration/ System System Memory Memory 115 GB/s 115 GB/s POWER8 POWER8 CPU CPU NVLink NVLink 80 GB/s 80 GB/s P100 P100 P100 P100 GPU GPU GPU GPU GPU GPU GPU GPU Memory Memory Memory Memory GPU PCIe CPU 16 GB/s System bottleneck Graphics System Memory Memory IBM aDVantage: data communication and GPU performance POWER8 + 78 ms Tesla P100+NVLink x86 baseD 170 ms GPU system ImageNet / Alexnet: Minibatch size = 128 ADD: Coherent Accelerator Processor Interface (CAPI) FPGA CAPP PCIe POWER8 Processor ...FPGAs, networking, memory... Typical I/O MoDel Flow Copy or Pin MMIO Notify Poll / Int Copy or Unpin Ret. From DD DD Call Acceleration Source Data Accelerator Completion Result Data Completion Flow with a Coherent MoDel ShareD Mem. ShareD Memory Acceleration Notify Accelerator Completion Focus on Enterprise Scale-Up Focus on Scale-Out and Enterprise Future Technology and Performance DriVen Cost and Acceleration DriVen Partner Chip POWER6 Architecture POWER7 Architecture POWER8 Architecture POWER9 Architecture POWER10 POWER8/9 2007 2008 2010 2012 2014 2016 2017 TBD 2018 - 20 2020+ POWER6 POWER6+ POWER7 POWER7+ POWER8 POWER8 P9 SO P9 SU P9 SO 2 cores 2 cores 8 cores 8 cores 12 cores w/ NVLink
    [Show full text]
  • IBM Power System E850 the Most Agile 4-Socket System in the Marketplace, Optimized for Performance, Reliability and Expansion
    IBM Systems Data Sheet IBM Power System E850 The most agile 4-socket system in the marketplace, optimized for performance, reliability and expansion Businesses today are demanding faster insights that analyze more data in Highlights new ways. They need to implement applications in days versus months, and they need to achieve all these goals while reducing IT costs. This is ●● ●●Designed for data and analytics, delivers creating new demands on IT infrastructures, requiring new levels of per- secure, reliable performance in a compact, 4-socket system formance and the flexibility to respond to new business opportunities, all at an affordable price. ●● ●●Can flexibly scale to rapidly respond to changing business needs The IBM® Power® System E850 server offers a unique blend of ●● ●●Can reduce IT costs through application enterprise-class capabilities in a space-efficient, 4-socket system with consolidation, higher availability and excellent price performance. With up to 48 IBM POWER8™ processor virtualization to yield over 70 percent utilization cores, advanced IBM PowerVM® virtualization that can yield over 70 percent system utilization and Capacity on Demand (CoD), no other 4-socket system in the industry delivers this combination of performance, efficiency and business agility. These capabilities make the Power E850 server an ideal platform for medium-size businesses and as a departmental server or data center building block for large enterprises. Designed for the demands of big data and analytics Businesses are amassing a wealth of data and IBM Power Systems™, built with innovation to support today’s data demands, can store it, secure it and, most important, extract actionable insight from it.
    [Show full text]
  • IBM Power Systems Performance Report Apr 13, 2021
    IBM Power Performance Report Power7 to Power10 September 8, 2021 Table of Contents 3 Introduction to Performance of IBM UNIX, IBM i, and Linux Operating System Servers 4 Section 1 – SPEC® CPU Benchmark Performance 4 Section 1a – Linux Multi-user SPEC® CPU2017 Performance (Power10) 4 Section 1b – Linux Multi-user SPEC® CPU2017 Performance (Power9) 4 Section 1c – AIX Multi-user SPEC® CPU2006 Performance (Power7, Power7+, Power8) 5 Section 1d – Linux Multi-user SPEC® CPU2006 Performance (Power7, Power7+, Power8) 6 Section 2 – AIX Multi-user Performance (rPerf) 6 Section 2a – AIX Multi-user Performance (Power8, Power9 and Power10) 9 Section 2b – AIX Multi-user Performance (Power9) in Non-default Processor Power Mode Setting 9 Section 2c – AIX Multi-user Performance (Power7 and Power7+) 13 Section 2d – AIX Capacity Upgrade on Demand Relative Performance Guidelines (Power8) 15 Section 2e – AIX Capacity Upgrade on Demand Relative Performance Guidelines (Power7 and Power7+) 20 Section 3 – CPW Benchmark Performance 19 Section 3a – CPW Benchmark Performance (Power8, Power9 and Power10) 22 Section 3b – CPW Benchmark Performance (Power7 and Power7+) 25 Section 4 – SPECjbb®2015 Benchmark Performance 25 Section 4a – SPECjbb®2015 Benchmark Performance (Power9) 25 Section 4b – SPECjbb®2015 Benchmark Performance (Power8) 25 Section 5 – AIX SAP® Standard Application Benchmark Performance 25 Section 5a – SAP® Sales and Distribution (SD) 2-Tier – AIX (Power7 to Power8) 26 Section 5b – SAP® Sales and Distribution (SD) 2-Tier – Linux on Power (Power7 to Power7+)
    [Show full text]
  • IBM Energyscale for POWER8 Processor-Based Systems
    IBM EnergyScale for POWER8 Processor-Based Systems November 2015 Martha Broyles Christopher J. Cain Todd Rosedahl Guillermo J. Silva Table of Contents Executive Overview...................................................................................................................................4 EnergyScale Features.................................................................................................................................5 Power Trending................................................................................................................................................................5 Thermal Reporting...........................................................................................................................................................5 Fixed Maximum Frequency Mode...................................................................................................................................6 Static Power Saver Mode.................................................................................................................................................6 Fixed Frequency Override...............................................................................................................................................6 Dynamic Power Saver Mode...........................................................................................................................................7 Power Management's Effect on System Performance................................................................................................7
    [Show full text]
  • IBM Power System S814 and S824 Technical Overview and Introduction
    Front cover IBM Power Systems S814 and S824 Technical Overview and Introduction Outstanding performance based on POWER8 processor technology 4U scale-out desktop and rack-mount servers Improved reliability, availability, and serviceability features Alexandre Bicas Caldeira Bartłomiej Grabowski Volker Haug Marc-Eric Kahle Cesar Diniz Maciel Monica Sanchez ibm.com/redbooks Redpaper International Technical Support Organization IBM Power System S814 and S824 Technical Overview and Introduction August 2014 REDP-5097-00 Note: Before using this information and the product it supports, read the information in “Notices” on page vii. First Edition (August 2014) This edition applies to IBM Power System S814 (8286-41A) and IBM Power System S824 (8286-42A) systems. © Copyright International Business Machines Corporation 2014. All rights reserved. Note to U.S. Government Users Restricted Rights -- Use, duplication or disclosure restricted by GSA ADP Schedule Contract with IBM Corp. Contents Notices . vii Trademarks . viii Preface . ix Authors. ix Now you can become a published author, too! . xi Comments welcome. xi Stay connected to IBM Redbooks . xi Chapter 1. General description . 1 1.1 Systems overview . 2 1.1.1 Power S814 server . 2 1.1.2 Power S824 server . 3 1.2 Operating environment . 3 1.3 Physical package . 4 1.3.1 Tower model . 4 1.3.2 Rack-mount model . 5 1.4 System features . 7 1.4.1 Power S814 system features . 7 1.4.2 Power S824 system features . 8 1.4.3 Minimum features . 9 1.4.4 Power supply features . 9 1.4.5 Processor module features . 9 1.4.6 Memory features .
    [Show full text]
  • Upgrade to POWER9 Planning Checklist
    Achieve your full business potential with IBM POWER9 Future-forward infrastructure designed to crush data-intensive workloads Upgrade to POWER9 Planning Checklist Using this checklist will help to ensure that your infrastructure strategy is aligned to your needs, avoiding potential cost overruns or capability shortfalls. 1 Determine current and future capacity 5 Identify all dependencies for major database requirements. Bring your team together, platforms, including Oracle, DB2, SAP HANA, assess your current application workload and open-source databases like EnterpriseDB, requirements and three- to five-year MongoDB, neo4j, and Redis. You’re likely outlook. You’ll then have a good picture running major databases on the Power of when and where application growth Systems platform; co-locating your current will take place, enabling you to secure servers may be a way to reduce expenditure capacity at the appropriate time on an and increase flexibility. as-needed basis. 6 Understand current and future data center 2 Assess operational efficiencies and identify environmental requirements. You may be opportunities to improve service levels unnecessarily overspending on power, while decreasing exposure to security and cooling and space. Savings here will help your compliancy issues/problems. With new organization avoid costs associated with data technologies that allow you to easily adjust center expansion. capacity, you will be in a much better position to lower costs, improve service Identify the requirements of your strategy for levels, and increase efficiency. 7 on- and off-premises cloud infrastructure. As you move to the cloud, ensure you 3 Create a detailed inventory of servers have a strong strategy to determine which across your entire IT infrastructure.
    [Show full text]
  • IBM POWER8 High-Performance Computing Guide: IBM Power System S822LC (8335-GTB) Edition
    Front cover IBM POWER8 High-Performance Computing Guide IBM Power System S822LC (8335-GTB) Edition Dino Quintero Joseph Apuzzo John Dunham Mauricio Faria de Oliveira Markus Hilger Desnes Augusto Nunes Rosario Wainer dos Santos Moschetta Alexander Pozdneev Redbooks International Technical Support Organization IBM POWER8 High-Performance Computing Guide: IBM Power System S822LC (8335-GTB) Edition May 2017 SG24-8371-00 Note: Before using this information and the product it supports, read the information in “Notices” on page ix. First Edition (May 2017) This edition applies to: IBM Platform LSF Standard 10.1.0.1 IBM XL Fortran v15.1.4 and v15.1.5 compilers IBM XLC/C++ v13.1.2 and v13.1.5 compilers IBM PE Developer Edition version 2.3 Red Hat Enterprise Linux (RHEL) 7.2 and 7.3 in little-endian mode © Copyright International Business Machines Corporation 2017. All rights reserved. Note to U.S. Government Users Restricted Rights -- Use, duplication or disclosure restricted by GSA ADP Schedule Contract with IBM Corp. Contents Notices . ix Trademarks . .x Preface . xi Authors. xi Now you can become a published author, too! . xiii Comments welcome. xiv Stay connected to IBM Redbooks . xiv Chapter 1. IBM Power System S822LC for HPC server overview . 1 1.1 IBM Power System S822LC for HPC server. 2 1.1.1 IBM POWER8 processor . 3 1.1.2 NVLink . 4 1.2 HPC system hardware components . 5 1.2.1 Login nodes . 6 1.2.2 Management nodes . 6 1.2.3 Compute nodes. 7 1.2.4 Compute racks . 7 1.2.5 High-performance interconnect.
    [Show full text]
  • POWER8: the First Openpower Processor
    POWER8: The first OpenPOWER processor Dr. Michael Gschwind Senior Technical Staff Member & Senior Manager IBM Power Systems #OpenPOWERSummit Join the conversation at #OpenPOWERSummit 1 OpenPOWER is about choice in large-scale data centers The choice to The choice to The choice to differentiate innovate grow . build workload • collaborative • delivered system optimized innovation in open performance solutions ecosystem • new capabilities . use best-of- • with open instead of breed interfaces technology scaling components from an open ecosystem Join the conversation at #OpenPOWERSummit Why Power and Why Now? . Power is optimized for server workloads . Power8 was optimized to simplify application porting . Power8 includes CAPI, the Coherent Accelerator Processor Interconnect • Building on a long history of IBM workload acceleration Join the conversation at #OpenPOWERSummit POWER8 Processor Cores • 12 cores (SMT8) 96 threads per chip • 2X internal data flows/queues • 64K data cache, 32K instruction cache Caches • 512 KB SRAM L2 / core • 96 MB eDRAM shared L3 • Up to 128 MB eDRAM L4 (off-chip) Accelerators • Crypto & memory expansion • Transactional Memory • VMM assist • Data Move / VM Mobility • Coherent Accelerator Processor Interface (CAPI) Join the conversation at #OpenPOWERSummit 4 POWER8 Core •Up to eight hardware threads per core (SMT8) •8 dispatch •10 issue •16 execution pipes: •2 FXU, 2 LSU, 2 LU, 4 FPU, 2 VMX, 1 Crypto, 1 DFU, 1 CR, 1 BR •Larger Issue queues (4 x 16-entry) •Larger global completion, Load/Store reorder queue •Improved branch prediction •Improved unaligned storage access •Improved data prefetch Join the conversation at #OpenPOWERSummit 5 POWER8 Architecture . High-performance LE support – Foundation for a new ecosystem . Organic application growth Power evolution – Instruction Fusion 1600 PowerPC .
    [Show full text]
  • IBM Power System S824L Technical Overview and Introduction
    Front cover IBM Power System S824L Technical Overview and Introduction Linux server built on OpenPOWER technologies Ready for scientific, Java, and Big Data & Analytics workloads Dual NVIDIA GPU accelerators supported Alexandre Bicas Caldeira YoungHoon Cho James Cruickshank Bartłomiej Grabowski Volker Haug Andrew Laidlaw Seulgi Yoppy Sung ibm.com/redbooks Redpaper International Technical Support Organization IBM Power System S824L Technical Overview and Introduction December 2014 REDP-5139-00 Note: Before using this information and the product it supports, read the information in “Notices” on page vii. First Edition (December 2014) This edition applies to IBM Power System S824L (8247-42L) servers. © Copyright International Business Machines Corporation 2014. All rights reserved. Note to U.S. Government Users Restricted Rights -- Use, duplication or disclosure restricted by GSA ADP Schedule Contract with IBM Corp. Contents Notices . vii Trademarks . viii Preface . ix Authors. ix Now you can become a published author, too! . .x Comments welcome. xi Stay connected to IBM Redbooks . xi Chapter 1. General description . 1 1.1 Power S824L server . 2 1.2 NVIDIA Tesla GPU Accelerators. 3 1.2.1 NVIDIA CUDA. 5 1.3 Operating environment . 5 1.4 Physical package . 6 1.5 Server features . 7 1.5.1 Server features for a Power S824L with a NVIDIA GPU. 7 1.5.2 Server features for a Power S824L without a GPU. 8 1.5.3 Minimum features . 8 1.5.4 Power supply features . 8 1.5.5 Processor module features . 9 1.5.6 Memory features . 9 1.6 PCIe slots . 10 1.7 Disk and media features .
    [Show full text]
  • POWER8® Processor-Based Systems RAS Introduction to Power Systems™ Reliability, Availability, and Serviceability
    POWER8® Processor-Based Systems RAS Introduction to Power Systems™ Reliability, Availability, and Serviceability March 17th, 2016 IBM Server and Technology Group Daniel Henderson Senior Technical Staff Member, RAS Information in this document is intended to be generally descriptive of a family of processors and servers. It is not intended to describe particular offerings of any individual servers. Not all features are available in all countries. Information contained herein is subject to change at any time without notice. Trademarks, Copyrights, Notices and Acknowledgements Trademarks IBM, the IBM logo, and ibm.com are trademarks or registered trademarks of International Business Machines Corporation in the United States, other countries, or both. These and other IBM trademarked terms are marked on their first occurrence in this information with the appropriate symbol (® or ™), indicating US registered or common law trademarks owned by IBM at the time this information was published. Such trademarks may also be registered or common law trademarks in other countries. A current list of IBM trademarks is available on the Web at http://www.ibm.com/legal/copytrade.shtml The following terms are trademarks of the International Business Machines Corporation in the United States, other countries, or both: Active AIX® POWER® POWER Hypervisor™ Power Systems™ Power Systems Software™ Memory™ POWER6® POWER7® POWER7+™ POWER8™ PowerHA® PowerLinux™ PowerVM® System x® System z® POWER Hypervisor™ Additional Trademarks may be identified in the body of this document. The following terms are trademarks of other companies: Intel, Intel Xeon, Intel logo, Intel Inside logo, and Intel Centrino logo are trademarks or registered trademarks of Intel Corporation or its subsidiaries in the United States and other countries.
    [Show full text]
  • POWER8Ò Processor Technology
    IBM Power Systems 9 January 2018 IBM Power Systems Facts and Features: Enterprise and Scale-out Systems with POWER8Ò Processor Technology 1 IBM Power Systems TaBle of contents Page no. Notes 3 Why Power Systems 4 IBM Power System S812LC, S822LC for Commercial Computing, S822LC for HPC, 5 S822LC for Big Data and S821LC IBM Power System S812L and S822L 6 IBM Power System S824L 7 IBM Power System S814 and S824 8 IBM Power System S812 and S822 9 IBM Power System E850C 10 IBM Power System E850 11 IBM Power System E870C 12 IBM Power System E870 13 IBM Power System E880C 14 IBM Power System E880 15 System S Class System Unit Details 16-19 Enterprise System, E850 System Unit and E870/E880 Node & Control Unit Details 19-20 Server I/O Drawers & Attachment 21-22 Physical Planning Characteristics 23 Warranty / Installation 24 Scale-out LC Server Systems Software Support 25 System S Class Systems Software Support 26-27 Enterprise Systems Software Support 28 Performance Notes & More Information 29 IBMÒ Power Systems™ servers and IBM BladeCenter® blade servers using IBM POWER7® and POWER7+® processors are described in a separate Facts and Features report dated July 2013 (POB03022-USEN-28). IBM Power Systems servers and IBM BladeCenter® blade servers using IBM POWER6® and POWER6+™ processors are described in a separate Facts and Features report dated April 2010 (POB03004-USEN-14). 2 IBM Power Systems These notes apply to the description tables for the pages which follow: Y Standard / Supported Optional Optionally Available / Supported N/A or – n/a Not Available / Supported or Not Applicable SOD Statement of General Direction announced SLES SUSE Linux Enterprise Server RHEL Red Hat Enterprise Linux A CoD capabilities include: Capacity Upgrade on Demand option – permanent processor or memory activation, Elastic Capacity on Demand – temporary processor or memory activation by the day, Utility Capacity on Demand – temporary processor activation by the minute, and Trial Capacity on Demand.
    [Show full text]