Manuals / Intel / Computer Equipment / Power Supply

Intel 170 Servers, AS/400 RISC Server, 7xx Servers manual 1500 1000, Concurrent Virtual Tapes

Models: 7xx Servers 170 Servers AS/400 RISC Server

1 368

Download 368 pages 6.76 Kb

Page 258

15.14 Concurrent Virtual Tapes

NOTE: Virtual tape is reading and writing to the same DASD so the maximum throughput with our concurrent and parallel measurements is different than our tape drive tests where we were reading from DASD and writing to tape.

Concurrent Virtual Tape for Large File

570 16 way 128GB Memory

800 DASD units for Virtual Tape Drives

3500
3000
2500
2000
GB/HR
1500
1000
500
0	Restore
Save	Restore
	Operation

1 Virtual Tape Device

2 Virtual Tape Devices

3 Virtual Tape Devices

4 Virtual Tape Devices

5 Virtual Tape Devices

	Concurrent Virtual Tape for User Mix
		570 16 way 128GB Memory
1000	800 DASD units for Virtual Tape Drives
1000
800
600		1 Virtual Tape Device
GB/HR		2 Virtual Tape Devices
400		3 Virtual Tape Devices
400		4 Virtual Tape Devices
		4 Virtual Tape Devices
200		5 Virtual Tape Devices
200
0	Save	Restore
	Save	Restore
		Operation
IBM i 6.1 Performance Capabilities Reference - January/April 2008

© Copyright IBM Corp. 2008

Chapter 15. Save/Restore Performance

258

Image 258

Intel 170 Servers, AS/400 RISC Server, 7xx Servers manual 1500 1000, Concurrent Virtual Tapes

Contents

January/April/October Toolcat Mkttools GET AS4ADV36 Package Table of Contents 117 Performance Observations and Tips107 System Application Server Instance 119158 10.2 DB2 for i5/OS access with Odbc 155157 11.1199 Supported Backup Device Rates 243196 200IPL Performance 284 17.7 File Level Backup Performance 29219.6 User Pool Faulting Guidelines 310 19.7 285335 Switchable IASP’s 327 Geographic Mirroring 329334 What the Estimator is Not 335Special Notices Disclaimer NoticeIpds Purpose of this Document Introduction Page Interactive Indicators and Metrics ISeries and AS/400 Risc Server Model Performance BehaviorOverview Disclaimer and Remaining Sections 3 V5R3Server Model Behavior V4R5 V5R2 4 V5R2 and V5R1Page 100 Server Model Differences Custom Server Model Model 7xx and 9/98 Model Performance Highlights of Model 7xx ServersPerformance Highlights of Model 170 Servers Performance Highlights of Custom Server Models Additional Server ConsiderationsInteractive Utilization ExampleServer Dynamic Tuning SDT Page Affects of Server Dynamic Tuning Interactive/Server characteristics in the real world Managing Interactive CapacityIs Interactive really Interactive? Page Migration from Traditional Models Interval end time hour and minute Number of interactive transactions per hourUpgrade Considerations for Interactive Capacity 13.2 V5R1 DSD Performance Behavior Domino-Complementary Processing V5R1 Non-Domino ProcessingInteractive Processing Logical Partitioning on a Dedicated Server Running Linux on a Dedicated ServerPage Effect of Dasd Type on Batch Batch PerformanceEffect of CPU Speed on Batch Batch Workload DescriptionJob Priority Tuning Parameters for BatchExpert Cache Dynamic Priority SchedulingPage DB2 for i5/OS Performance New for i5/OS V6R1I5/OS V6R1 SQE Query Coverage Query Attribute POWER5 Processor POWER6 Processor DB2 i5/OS V5R4 Highlights Records Inserted/SelectedI5/OS V5R4 SQE Query Coverage I5/OS V5R3 Highlights I5/OS V5R3 SQE Query CoverageCommon table expressions Deletes Derived tables Like/Substring predicates DB2 Multisystem LOB columns I5/OS V5R3 SQE Performance EnhancementsPartitioned Table Support V5R2 Highlights Introduction of the SQL Query Engine Materialized Query Table Support Fast Delete SupportSQE Optimizer SQE Query Engine SQE Database StatisticsStatistic Usage Example Indexing SQE for V5R2 SummaryEncoded Vector Indices EVIs Decision Support Queries DB2 Symmetric Multiprocessing feature IntroductionDB2 for i5/OS Memory Sharing Considerations SMP Performance SummaryJournaling and Commitment Control JournalingRemote Journal Function System-Managed Access Path Protection Smapp SmappDB2 Multisystem for i5/OS Commitment ControlCopyright IBM Corp Referential Integrity TriggersVariable Length Fields DescriptionPerformance Expectations Reuse Deleted Record Space Description of FunctionComparison to Normal Inserts Performance References for DB2 Copyright IBM Corp Communications Performance System i Ethernet Solutions CCIN3Communication Performance Test Environment HardwareSoftware Communication and Storage observations Dasd PerformanceLarge data transfer FTP TCP/IP non-secure performance Streaming PerformanceRR & Acrr Performance TCP/IP Secure Performance249.7 53.4 48.0 31.3 27.4 SSL Performance1167 565.4 530.0 479.6 462.1 202.2 478.4 55.7 53.3 36.9 31.9SSL Relative Performance VPN Performance27.5 VPN Relative PerformancePerformance Observations and Tips 10.9 15.4 18.7 88.8Page APPC, ICF, CPI-C, and Anynet Page HPR and Enterprise extender considerations Test Type Hprip Link Hprip Link Speed AnyNetSpeed = 10Mbps = 100Mbps Additional Information For network information select NetworkingWeb Server and WebSphere Performance Http Server powered by Apache Relative Performance Metrics Http Server powered by Apache for i5/OS V5R4 Relative Capacity for CGI Http Server powered by Apache for i5/OS Page Page Page Page Zend Core for System ConfigurationPHP Zend Core for Database and Workload DescriptionDatabase Access OS / DBDatabase Persistent versus Non-Persistent Connections Database Isolation Levels Zend PlatformPHP System Sizing Thi WebSphere Application ServerWhat’s new in V6R1? ShoUsing ‘threadUsed=false custom property toolbox Jdbc Tuning changes in V6R1Using direct map native Jdbc Use of unix sockets toolbox JdbcPage Topology of the Trade Application Page WebSphere Application Server Trade3 Measurement Results Trade Capacity ResultsTrade Scalability Results Trade 6 PrimitivesDescription of Trade primitives in Figure WebSphere Trade 3 Primitives Accelerator for System Trade Jdbc Response Time 1 User Time seconds Page WebSphere Application Server V51 Express IBM WebFacing WebFaced applications, one simple, oneModerate, and one complex. See .4, for Many 5250 applications have been implemented Version 5.0 of WebfacingMore in line with the other two machines With best performance techniques, such asTuning the Record Definition Cache Display File Record I/O ProcessingDefinitions would be retained in the cache with few cache Cache Management Definition Cache Content ViewerCache Size Effect Misses for not commonly used recordsCache Viewer Button operations Button Operation Cache Management Record Definition LoaderRecord Definition Loader Button operations Operation CompressionEnabling Compression in IBM Http Server powered by Apache Webfacing CompressionAdditional Resources PartnerWorld for Developers Webfacing website WebSphere Host Access Transformation Services Hats What’s new with V5R4 and HatsHats Customization Hats Customization CPW/User System Application Server Instance IBM Systems Workload Estimator for HatsPage WebSphere Portal WebSphere CommerceConnect for iSeries WebSphere Commerce Payments Performance Tips and TechniquesConnect for iSeries Connector Types Performance Observations/TipsPage Java Performance IntroductionWhat’s new in V6R1 IBM Technology for Java 32-bit and 64-bit Native Code Garbage CollectionClassic VM 64-bit JIT CompilerPage Garbage Collection Bytecode Verification Determining Which JVM to Use Functional ConsiderationsPerformance Considerations Capacity Planning General GuidelinesJava Performance Tips and Techniques I5/OS Specific Java Tips and Techniques Classic VM-specific TipsConsider the special property os400.jit.mmi.threshold Java Language Performance TipsMinimize synchronized methods Instead, the above procedure should be written as Java i5/OS Database Access TipsPool Database Connections Resources IBM Hardware Solutions System i Cryptographic SolutionsIBM Software Solutions Cryptography PerformanceCryptography Performance Test Environment CSP API SetsSoftware Cryptographic API Performance Measurement ResultsHardware Cryptographic API Performance Bytes/Second4764 Cipher Encrypt PerformanceEncryption Threads Transactions/second Bytes/secondCryptography Observations, Tips and Recommendations CCA CSPSupported number of 4764 Cryptographic Coprocessors ISeries NetServer File Serving Performance ISeries NetServer File Serving PerformanceResponse Time Conclusion/ExplanationsPage 10.1 DB2 for i5/OS access with Jdbc Jdbc Performance Tuning TipsDB2 for i5/OS Jdbc and Odbc Performance References for Jdbc 10.2 DB2 for i5/OS access with Odbc Odbc Performance Settings Page References for Odbc Microsoft Odbc webpageDomino on Domino Workload Descriptions Mail and Calendaring Users MCUDomino Web Access formerly known as iNotes Web Access DominoDomino Domain Monitor Domino Web Access client improvements with Domino Additional memory was added for this testResponse Time and Megahertz relationship Collaboration Edition and Domino Edition offerings Collaboration EditionDomino Edition Performance Tips / Techniques Page Page Domino Web Access Domino Subsystem TuningPerformance Monitoring Statistics Main Storage OptionsMinimize Thousands Sizing Domino on System V5R3 Main Storage Options Response TimesLpar and Partial Processor Considerations Performance Data Collection ToolsSystem i NotesBench Audits and Benchmarks Queue Manager Start Following an Abnormal End WebSphere MQ for iSeriesPerformance Improvements for WebSphere MQ V5.3 CSD6 Checkpoint Following a Journal Receiver Roll-overTest Description and Results Conclusions, Recommendations and TipsOther Sources of Information Summary Basic Requirements -- Where Linux RunsLinux on iSeries Performance Key IdeasLinux on iSeries Technical Overview Linux on iSeries ArchitectureLinux on iSeries Run-time Support Basic Configuration and Performance Questions General Performance Information and Results Computational Performance -- C-based codeNumber Processors Partition Computational Performance -- JavaWeb Serving Performance Gcc and High Optimization gcc compiler option -O3 TCP/IP Function Megabit Ethernet LAN Virtual LANNetwork Operations Gcc Compiler, VersionValue of Virtual LAN and Virtual Disk Virtual Disk13.7 DB2 UDB for Linux on iSeries Linux on iSeries and IBM eServer Workload Estimator Top Tips for Linux on iSeries PerformancePage Page Page Dasd Performance Internal Native AttachmentDirect Attach Native Hardware Characteristics Devices & Controllers14.1.2 iV5R2 Direct Attach Dasd Intensive Workload Performance ComparisonIntensive Workload 14.1.3 571B 14.1.3.1 571B RAID5 vs RAID6 10 15K 35GB Dasd14.1.3.2 571B IOP vs Iopless 10 15K 35GB Dasd 14.1.4 571B, 5709, 573D, 5703, 2780 IOA Comparison Chart Dasd14.1.4.4 14.1.5.1 Comparing 571E/574F and 571F/575B IOP and IOPLess 14.1.6.21000 2000 3000 4000 Ops/Sec 14.1.7.4 Performance Limits on the 571F/575B 14.1.8.1Page 14.1.9.2 14.1.9.1 Direct Attach 571E/574F and 571F/575B Observations New in iV5R4M5 14.2.1 9406-MMA CEC vs 9406-570 CEC Dasd 100 200 300 400 500 600 Ops/SecRAID Hot Spare 14.2.3 12X Loop Testing New in iV6R1M0 Encrypted ASP Non Encrypted ASP vs Encrypted ASPNon Encrypted ASP vs Encrypted ASP 14.3.2 57B8/57B7 IOA 100 350 600 850 1100 1350 1600 1850POWER6 520 57B8/57B7 6 RAID5 Dasd in CEC 14.3.3 572A IOA Page SAN Storage Area Network External 14.5 iV6R1M0 -- Vios and IVM Considerations General Vios Considerations Generic ConceptsGeneric Configuration Concepts Page Page Chdev -dev hdisk03 -attr pv=yes will assign a Pvid to hdisk3 Page Vios and JS12 Express and JS22 Express Considerations 001 10000 20000 30000 40000 50000 60000 Vios CPU Vios Average Dasd Response Time Seconds BladeCenter S and JS12 Express Page Page System Workload Response Time Seconds IBM i operating system 5.4 Virtual Scsi Performance Page Introduction Virtual Scsi Performance Examples Native vs. Virtual Performance Virtual Scsi Bandwidth-Multiple Network Storage SpacesNwsd Read Scaling Virtual Scsi Bandwidth-Disk Scaling Sizing when using Dedicated Processors SizingCPU milliseconds to process virtual Scsi I/O transaction Sizing when using Micro-Partitioning Sizing memory Disk 1GB Sequential ReadAIX Virtual IO Client Performance Guide Save/Restore Performance Supported Backup Device RatesData Compression Dtacpr Save Command Parameters that Affect PerformanceUse Optimum Block Size Useoptblk Data Compaction CompactUser Mix WorkloadsDatabase File related Workloads User Mix 3GB contains 12,300 objectsComparing Performance Data Lower Performing Backup Devices Medium & High Performing Backup DevicesUltra High Performing Backup Devices Use of Multiple Backup Devices Parallel and Concurrent Library Measurements Save and Restore Rates Large File ConcurrentSave and Restore Rates Large File ParallelUser Mix Concurrent Runs User Mix Concurrent450 440 430 420 Number of Processors Affect PerformanceDasd and Backup Devices Sharing a Tower 100200 Virtual TapeData To Save Hours 500 Parallel Virtual Tapes1500 1000 Concurrent Virtual TapesSave and Restore Scaling using a Virtual Tape Drive Large File Virtual Tape Scaling Save Write to Virtual tapeLarge File Save User Mix Saves High-End Tape Placement on System PCI-X ™ In a 5094/5294 tower use slot C08 or C09700 250 15.19 5XX Tape Device Rates SLR60Measurements in GB/HR all 8 Dasd in the system ASP Source File 1GB 15.21 5XX DVD RAM and Optical Library Software Compression High15.23 9406-MMA DVD RAM ASP15.24 9406-MMA 576B IOPLess IOA 1GBIV5R4 January What’s New and Tips on Performance What’s New IV6R1M0 MarchIV5R4M5 July AugustIPL Performance IPL Performance ConsiderationsIPL Test Description Software Configuration 16.3 9406-MMA System Hardware InformationActive Database 16.5 9406-MMA IPL Performance Measurements Abnormal 1 iV5R4M5 Normal IPL Power-On Cold Start16.4 9406-MMA IPL Performance Measurements Normal 1 iV5R4M5 Abnormal IPL Partition MSDMSD Affects on IPL Performance Measurements 16.7 5XX System Hardware Information 16.7.1 5XX Small system Hardware Configuration16.7.2 5XX Large system Hardware Configuration Database16.9 5XX IPL Performance Measurements Abnormal 1 Normal IPL Power-On Cold Start16.8 5XX IPL Performance Measurements Normal Measurement units are in hours, minutes and seconds16.10 5XX IOP vs IOPLess effects on IPL Performance Normal IPL TipsIntegrated BladeCenter and System x Performance V5R4 iSCSI Host Bus Adapter iSCSI HBAIntegrated xSeries Adapter IXA Effects of Windows and Linux loads on the host system Integrated xSeries Servers IXS17.2.1 IXS/IXA Disk I/O Operations Extended Write Operations ISCSI Disk I/O OperationsISCSI virtual I/O private memory pool Crtsbsd SBSDQGPL/QFPHIS POOLS1 10000Virtual Ethernet Connections 17.2.5 IXS/IXA IOP ResourceIXS and IXA attached servers ISCSI attached serversDisk I/O CPU Cost Machine Pool MBytes Base Pool Qfphis Private Pool TotalFurther notes about IXS/IXA Disk Operations CPW per 1k Disk OperationsDisk I/O Throughput ISCSI Target, IXA Capacity Comparison MB per SecondVirtual Ethernet CPU Cost and Capacities VE Capacity ComparisonsVE CPW Cost PP TCP Stream i5/OS to WindowsFile Level Backup Performance Windows CPU CostAdditional Sources of Information Flbu SAV / RST RatesMicrosoft Hardware Compatibility Test URL See V5R2 Additions Logical Partitioning LparV5R3 Information General TipsConsiderations V5R1 AdditionsLpar Performance Considerations Performance on a 12-way systemPage Lpar Throughput Increase Lpar Measurements Over a Standalone 12-way 5920 26 %5700 21 % Miscellaneous Performance Information TPC-C Commercial PerformanceSAP Performance Information SPECjbb2000 VolanoMarkDelay Cost Terminology Dynamic Priority SchedulingPriority Mapping to Delay Cost Curves Performance Testing Results Qdynptyscd = ‘1’ on Qdynptyscd = ‘0’Memory Tuning Using the Qpfradj System Value Main Storage Sizing GuidelinesAdditional Memory Tuning Techniques User Pool Faulting Guidelines Setobjacc Set Object AccessLarge Memory Systems Interactive Batch19.7 AS/400 NetFinity Capacity Planning AS/400 NetFinity Software Inventory Performance Conclusions/Recommendations for NetFinity History Adjusting Your Performance Tuning for ThreadsGeneral Performance Tips and Techniques Coming ChangeProblem SolutionGeneral Performance Guidelines -- Effects of Compilation In-liningOptimization Levels Theory -- and Practice System Level Considerations Typical Storage CostsOrder1 Orderj Ordert OrderN Brief ExampleWhich is more important? Short but Important Tip about Data BaseFinal Thought About Memory and Competitiveness Hardware Multi-threading HMT HMT DescribedSome key similarities and differences are HMT Feature SMT FeatureHMT and SMT Compared and Contrasted Models With/Without HMTPOWER6 520 Memory Considerations Aligning Floating Point Data on Power6 Page High Availability Performance High Availability Switchable Resources ConsiderationsSwitchable IASP’s Workload DescriptionHardware Configuration Workload ConfigurationSystem Configuration Cabling MapSwitchover Tips Geographic MirroringWorkload Description Large System Configuration Hardware Configuration Cabling MapSynchronization on an idle system Switchable Towers using Geographic MirroringGeographic Mirroring Measurements Active StateGeographic Mirroring Tips IBM Systems Workload Estimator Estimator Access Merging PM for System i data into the EstimatorWhat the Estimator is Not Page Appendix A. CPW and CIW Descriptions Commercial Processing Workload CPWCPW Application Description Compute Intensive Workload CIW Page Page Performance Data Collection Services How Collection Services worksStarting Collection Services Batch Modeling Tool Bchmdl Page Page CPW values for the IBM Power Systems IBM i operating system V6R1 Additions OctoberCPW values for the IBM Power 595 IBM i operating system V6R1 Additions AugustV6R1 Additions April Table C.3.1. CPW values for Power System Models V6R1 Additions JanuaryV5R4 Additions July IBM System i using the POWER6 processor technologyIBM System i using the POWER5 processor technology V5R4 Additions January/May/August 2006 and January/AprilModel Edition Accelerator Value V5R3 Additions May, July, August, October 2004, JulyIBM ~ i5 Servers ExpressCPU 750 V5R2 Additions February, May, JulyISeries Model 8xx Servers 540Models share L2 cache between 2 processors Model 810 and 825 iSeries for Domino FebruaryV5R2 Additions Standard Models 8xx Servers10 V5R1 Additions Table C.9.2.1 Standard Models 8xx ServersModel 8xx Servers Table C.9.1.1 Model 8xx ServersTable C.10.2.1 Model 2xx Servers Model 2xx Servers10.3 V5R1 Dedicated Server for Domino Table C.10.3 .1 Dedicated Servers for DominoCapacity Upgrade on-demand Models CPW Values and Interactive Features for CUoD ModelsActive Stand-by 840-2352 840-2353 840-2354 Table C.10.4.1.1 V5R1 Capacity Upgrade on-demand Models 1650011 V4R5 Additions 11.1 AS/400e Model 8xx ServersTable C.11.1 Model 8xx Servers All new Condor models Dedicated Server for Domino Table C.11.2.1 Model 2xx ServersTable C.11.3.1 Dedicated Server for Domino 12.1 AS/400e Model 7xx Servers SB Models12 V4R4 Additions Table C.11.4.1 SB ModelsModel 170 Servers Current 170 Servers Table C.12.2.1 Current Model 170 Servers2407 2408 2409 120 AS/400e Dedicated Server for DominoPrevious Model 170 Servers Table C.12.2.2 Dedicated Server for Domino15 AS/400 Advanced Servers 13 AS/400e Model Sxx Servers14 AS/400e Custom Servers Table C.13.1 AS/400e Servers16 AS/400e Custom Application Server Model SB1 Table C.15.1 AS/400 Advanced Servers V4R1 and V4R2Table C.15.2 AS/400 Advanced Servers V3R7 Table C.17.1 AS/400 Risc Systems 17 AS/400 Models 4xx, 5xx and 6xx SystemsTable C.16.1 AS/400e Custom Application Server Model SB1 Table C.17.2 AS/400e Systems18 AS/400 Cisc Model Capacities Table C.18.6 AS/400 Cisc Model 9406 Systems Table C.18.7 AS/400 Advanced Systems CiscTable C.18.8 AS/400 Advanced Servers Cisc