Considerations for large-scale SharePoint deployments on SQL Server

Name: Joel OlesonTitle: Sr. Tech Prod MgrCompany: Quest Software

http://www.facebook.com/photo.php?pid=2607453&id=613898782

Audience PollNew to SharePoint?SQL Admins?Large-scale Implementation (>1TB) experience?Scalability or performance issues in SharePoint deployments?

Session Overview Lightweight

Understanding SharePoint databasesSQL PerformanceSQL Server 2008 with SharePoint

HeavyweightArchitectural Design ConsiderationsReal-world scenarios

Business RequirementsLogical and Physical ArchitectureArchitectural Design Statistical Results

Appendix: DB Sizes, Content Distribution…

=Lightweight

Real World ExamplesInformation based on real-world, large-scale SharePoint Implementations.

Large software company (Microsoft)Intranet Portal for 120K usersGlobal Enterprise Collaboration Solution (~20TB)Scalable Hosting Solution (SharePoint Online)

Large automotive manufacturerLoan Origination Application / Document Repository~50 Million content items (~6 TB)

Understanding the SharePoint Databases

Farm

• Config• Servers• Web Apps• Solutions• Global

Config

Web App

• Content 1..2• Site

Collections• Sites• Lists• Pages• Documents• DWPs

SSP

• Search• Properties

• SSP• My site

host config• Profiles• BDC config• Excel Calc

Disk I/O Demand

Search

Most Demand

Config

Medium Demand

+SSP

*Content..

Low Demand

* Except during backup and Indexing + Except during Profile Import

Temp

Master

Model

Tlogs

Top Performance Killers

1. Indexing/Crawling2. Backup (SQL & Tape)3. Profile Import4. Misc Timer Jobs – User Sync for large #s of

Users5. STSADM Backup/Restore6. Large List Operations7. Heavy User Operation List Import/Write

Content Db

Content Database

Sites WebsDoc

Stream

Config

Config Database

Sites Servers VServers

SQL Server 2008 with Windows Server 2008Transactional Performance with SQL Server 2008 Dramatically outperformed SQL 2005 on Win 2003.Compressed backup in the boxSupport for SQL External Blob StorageIncreased resiliencyTransparent Encryption

See Performance Gains athttp://msdn.microsoft.com/en-us/library/dd263442.aspx

=heavyweight

Architectural Design ConsiderationsDatabase Volumes

Separate database volumes into unique LUN’s consisting of unique physical disk spindles.In a heavily read-oriented internet (portal) site, prioritize data over logs.Separate out Search database transaction log from content database transaction logs.

Architectural Design ConsiderationsSQL TempDB Data Files

Optimal TempDB data file sizes can be calculated using the following formula:

[MAX DB SIZE (KB)] X [.25] / [# CORES] = DATA FILE SIZE (KB)

Calculation result (starting size) should be roughly equal to 25% of the largest content or search DB.Use RAID 10; separate LUN from other database objects (content, search, etc…).“Autogrow” feature set to a fixed amount; if auto grow occurs, permanently increase TempDB size.TempDB Log file separated to unique LUN.

Content Databases100 content databases per Web application

100GB per content databaseCAUTION: DB locking issues reported in collaborative DM scenarios above 100GB

Need to ensure that you understand the issues based on number of users, usage profiles, etc…

Service Level Agreement (SLA) requirements for backup and restore will also have an impact on this decision.

Architectural Design Considerations

Content Databases - ContinuedPre-construct and pre-sizeUse RAID 5 or RAID 10 logical units

RAID 10 is the best choice when cost is not a concern. RAID 5 will be sufficient and will save on costs, since content databases tend to be more read intensive than write intensive.

Multi-core computer running SQL ServerPrimary file group could consist of a data file for each CPU core present in SQL Server.

Architectural Design Considerations

Database MaintenanceSQL Server SP2 is needed if using the DB maintenance wizard (KB930887).Plan regular defrag of databases

Performance - Average Disk Queue LengthSingle Digit values are optimal.Occasional double-digit values aren’t a large concern.Sustained triple-digit values require attention.

Architectural Design Considerations

PerformanceThe recommended practice for separating the database volume types for the transaction log files to unique LUN’s follows.

Content Database Log Files.Search Database Log Files.

Consider filegroups for search database

Architectural Design Considerations

Architectural Design ConsiderationsTopology

A single list should not have more than 2,000 items per list container.

A container represents the root of the list, as well as any folders within the list; a folder is a container because other list items are stored within it.Whitepaper: Working with large lists in Office SharePoint Server 2007 (Steve Peschka)

http://go.microsoft.com/fwlink/?LinkId=95450

Disk Drive Speed15K RPM recommended.

IIS Application PoolsEnsure “Max Used Memory” setting utilizes all the available RAM in your WFE’s.

Architectural Design ConsiderationsSTSADM Command-line Tool and CreateSiteInNewDB OperationGary Lapoint STSADM Extensions for Site Collection DB maintenanceCodeplex.com/governance tools for archive & delete capture

Large Scale Manufacture

Real-world ScenariosAutomotive Mfgr. Business Requirements (Phase I)

Loan Origination Application built on Office SharePoint Server 2007Ability to manage 10.5 million images.System performance with a “normal” input load defined as receipt of 27,000 images per business day = 10 hours.Simulate user load to represent 200 users for search, view & update with 2x peak

Real-world ScenariosData Load Process (Phase I)

Used KnowledgeLake Document Release EngineLoaded 9.17 documents/second per server Employs a high-volume, storage-based folder architecture within SharePoint to ensure UI responsiveness.

Executed on 4 servers. Using this application, we were able to achieve:

An average document load throughput of 36.6 documents per second!An average daily input of 3.17 million documents!

10.5 million documents with only 28% utilization!

Real-world ScenariosData Load Process (Phase II)

15 million documents consisting of Word (.docx), Excel (.xlsx), PowerPoint (.pptx) and Adobe PDF.Five Web Front-Ends were used for the load process.Peak Load Rate:

24.3 docs per second/2.1 million documents per day.Average Load Rate:

~1.9 million documents per day.Load Time:

8 days.NOTE: Load rates included automation process that created the PDF files.

What does the logical architecture look like?!

What does the physical architecture look like?!

Scale OUT…

Scale UP…

What does the site topology look like?!

Phase I17 Divisional Site Collections / DB’sPhase II10 Departmental Site Collections / DB’s

What does the storage architecture look like?

Database SizesPhase I

10.00

10.00

20.00

30.00

40.00

50.00

60.00

70.00

80.00

Div01Div02Div03Div04Div05Div06Div07Div08Div09Div10Div11Div12Div13Div14Div15Div16Div17

Architectural Design Statistical ResultsPhase I

Designed Once / Built OnceNo architecture OR configuration changes were required after the initial build was completed.10.5+ million documents loaded into the system in approximately 60 hours!Full Crawl indexed 10 Million items in 32 hours!Average content database size for divisional breakouts was 60GB

Architectural Design Statistical ResultsPhase II

Search database size was 539GB.Lesson Learned: Large search database caused disk I/O contention; break this out into multiple data file allocations matching the number of core processors on SQL Server, and spread them over unique LUN’s.

Total Index size was 162GB!Average Content database size for Divisional breakouts was 200.65GB!Average Content database size for Departmental breakouts was 137.60GB!

Large Scale Pharma

Real-world ScenariosPharmaceutical Business Requirements

Collaboration Portal built on Office SharePoint Server 2007Validate ~40TB of content storage.Identify performance characteristics and provide guidance around content database sizingFAST search integration

Real-world ScenariosData Load Process

71,524,357 documents loaded across two SharePoint Farms 10.92 days!Content was spread across the farms into 165 unique content databases.

6,240 Site Collections, each containing 10 sub-sites for a total of 62,400 sites.Database sizes were pre-configured to vary in size from 100GB to 350GB to determine performance and/or SLA impacts.

What does the logical architecture look like?!

What does the physical architecture look like?!

What does the site topology look like?!

165 Content DB’s6,240 Site Collections

10 Sub-Sites in each collection:62,400 Sites!

What does the storage architecture look like?

Architectural Design Statistical ResultsConclusion

User LoadsStress tests included 2 - 3,000 concurrent users.

Based on the 10% rule, testing completed equated to an environment representing 300,000 users!

RAW number of RPS during peak times is 1,469 at Pharma.

773 RPS, which equates to 346.59 ACTUAL RPS!

FAST Search IntegrationSuccessfully integrated FAST search capabilities, indexed content corpus and served search results as expected.

Large-Scale Case Study AvailableSharePoint Scalability and Performance Whitepaper

Contains majority of content you will see here, along with test results you won’t see here.TechNet topic: http://go.microsoft.com/fwlink/?LinkId=120901Word 2007 format: http://go.microsoft.com/fwlink/?LinkId=120881Word 2000-2003 format: http://go.microsoft.com/fwlink/?LinkId=120890 PDF format: http://go.microsoft.com/fwlink/?LinkId=120891

question & answer

Appendix

Database SizesMPSC/Nissan Phase I

Type DB Name Volume Size (GB)Search MPSC_SharedServices_Search_DB.mdf SearchDb_Vol(G:) 63.40Divisional Content WSS_Content_MPSC1_Div01.mdf Content1_Vol(H:) 57.00Divisional Content WSS_Content_MPSC1_Div02.mdf Content2_Vol(I:) 60.70Divisional Content WSS_Content_MPSC1_Div03.mdf Content3_Vol(J:) 72.00Divisional Content WSS_Content_MPSC1_Div04.mdf Content4_Vol(K:) 60.10Divisional Content WSS_Content_MPSC1_Div05.mdf Content5_Vol(L:) 23.90Divisional Content WSS_Content_MPSC1_Div06.mdf Content6_Vol(M:) 60.60Divisional Content WSS_Content_MPSC1_Div07.mdf Content7_Vol(N:) 72.70Divisional Content WSS_Content_MPSC1_Div08.mdf Content8_Vol(O:) 69.80Divisional Content WSS_Content_MPSC1_Div09.mdf Content1_Vol(H:) 35.50Divisional Content WSS_Content_MPSC1_Div10.mdf Content2_Vol(I:) 65.60Divisional Content WSS_Content_MPSC1_Div11.mdf Content3_Vol(J:) 61.50Divisional Content WSS_Content_MPSC1_Div12.mdf Content4_Vol(K:) 65.60Divisional Content WSS_Content_MPSC1_Div13.mdf Content5_Vol(L:) 77.70Divisional Content WSS_Content_MPSC1_Div14.mdf Content6_Vol(M:) 64.90Divisional Content WSS_Content_MPSC1_Div15.mdf Content7_Vol(N:) 25.90Divisional Content WSS_Content_MPSC1_Div16.mdf Content8_Vol(O:) 65.50Divisional Content WSS_Content_MPSC1_Div17.mdf Content1_Vol(H:) 78.30 TOTAL STORAGE SIZE: 1,080.70 TOTAL DIVISIONAL CONTENT DB SIZE: 1,017.30 AVERAGE DIVISIONAL CONTENT DB SIZE: 59.84

Database SizesMPSC/Nissan Phase II

DB Type DB Name Volume Size (GB) # #Search MPSC_SharedServices_Search_DB.mdf SearchDb_Vol(G:) 539.00 ITEMS FOLDERS TIFF TIFFDivisional Content WSS_Content_MPSC1_Div01.mdf Content1_Vol(H:) 191.00 2,256,639 13,250Divisional Content WSS_Content_MPSC1_Div02.mdf Content2_Vol(I:) 198.00 2,206,762 13,244Divisional Content WSS_Content_MPSC1_Div03.mdf Content3_Vol(J:) 201.00 1,877,409 12,295Divisional Content WSS_Content_MPSC1_Div04.mdf Content4_Vol(K:) 197.00 2,205,250 13,243Divisional Content WSS_Content_MPSC1_Div05.mdf Content5_Vol(L:) 201.00 1,847,333 13,086Divisional Content WSS_Content_MPSC1_Div06.mdf Content6_Vol(M:) 199.00 2,205,712 13,241Divisional Content WSS_Content_MPSC1_Div07.mdf Content7_Vol(N:) 199.00 1,851,289 12,097Divisional Content WSS_Content_MPSC1_Div08.mdf Content8_Vol(O:) 201.00 1,938,231 12,642Divisional Content WSS_Content_MPSC1_Div09.mdf Content1_Vol(H:) 188.00 1,787,785 13,216Divisional Content WSS_Content_MPSC1_Div10.mdf Content2_Vol(I:) 199.00 2,179,837 11,828Divisional Content WSS_Content_MPSC1_Div11.mdf Content3_Vol(J:) 200.00 2,293,663 12,015Divisional Content WSS_Content_MPSC1_Div12.mdf Content4_Vol(K:) 203.00 2,216,798 12,016Divisional Content WSS_Content_MPSC1_Div13.mdf Content5_Vol(L:) 200.00 1,858,731 11,254Divisional Content WSS_Content_MPSC1_Div14.mdf Content6_Vol(M:) 202.00 2,224,727 12,016Divisional Content WSS_Content_MPSC1_Div15.mdf Content7_Vol(N:) 227.00 1,974,025 12,491Divisional Content WSS_Content_MPSC1_Div16.mdf Content8_Vol(O:) 203.00 2,214,368 12,016Divisional Content WSS_Content_MPSC1_Div17.mdf Content1_Vol(H:) 202.00 1,867,572 11,446 TOTAL DIVISIONAL CONTENT: 3,411.00 35,006,131 211,396 AVERAGE DIVISIONAL CONTENT: 200.65 2,059,184 12,435 Office/PDF Office/PDF docx pptx xlsx pdfDepartmental Content WSS_Content_MPSC2_Dpt01.mdf Content1_Vol(H:) 137.00 1,504,884 11,743 868,033 57,255 521,996 57,600 1,504,884Departmental Content WSS_Content_MPSC2_Dpt02.mdf Content2_Vol(I:) 138.00 1,507,553 11,600 869,546 57,616 522,878 57,513 1,507,553Departmental Content WSS_Content_MPSC2_Dpt03.mdf Content3_Vol(J:) 138.00 1,506,906 11,601 868,779 57,804 522,467 57,856 1,506,906Departmental Content WSS_Content_MPSC2_Dpt04.mdf Content4_Vol(K:) 137.00 1,505,731 11,601 868,455 57,536 522,215 57,525 1,505,731Departmental Content WSS_Content_MPSC2_Dpt05.mdf Content5_Vol(L:) 138.00 1,505,740 11,601 869,784 57,468 520,818 57,670 1,505,740Departmental Content WSS_Content_MPSC2_Dpt06.mdf Content6_Vol(M:) 138.00 1,507,524 11,601 869,765 57,765 522,241 57,753 1,507,524Departmental Content WSS_Content_MPSC2_Dpt07.mdf Content7_Vol(N:) 138.00 1,507,108 11,601 870,284 57,362 521,590 57,872 1,507,108Departmental Content WSS_Content_MPSC2_Dpt08.mdf Content8_Vol(O:) 137.00 1,503,853 11,601 867,015 57,318 521,748 57,772 1,503,853Departmental Content WSS_Content_MPSC2_Dpt09.mdf Content5_Vol(L:) 137.00 1,505,234 11,601 868,209 57,518 521,934 57,573 1,505,234Departmental Content WSS_Content_MPSC2_Dpt10.mdf Content7_Vol(N:) 138.00 1,508,146 11,599 870,315 57,571 522,393 57,867 1,508,146 TOTAL DEPARTMENTAL CONTENT: 1,376.00 15,062,679 116,149 8,690,185 575,213 5,220,280 577,001 AVERAGE DEPARTMENTAL CONTENT: 137.60 1,506,268 11,615 869,019 57,521 522,028 57,700 GRAND TOTAL CONTENT: 5,326.00 50,068,810 327,545 GRAND TOTAL AVERAGE CONTENT: 197.26 1,854,400 12,131

MPSC/Nissan Phase I14 individual performance tests were run to simulate various load scenarios.

Performance of Components Over Time

Load Averages

Search Load Modify Load Insert Load View Image Load Test Crawl Cycle WFE CPU Index CPU Target CPU SQL CPU TempDb Q TranLog Q SearchDb Q ContentDb Q Page Load TimeImage Load Time

Run # Cncnt. Users Test % Items/Min Count Minutes Minutes % % % % Length Length Length Length Seconds Seconds

1* 0 0 0 0 20 15 6.325 0.700 0.330 3.900 0.000 0.001 0.000 0.000 N/A N/A

2 200 0 0 0 20 15 19.350 0.700 0.320 6.500 0.000 0.001 0.000 0.000 0.042 N/A

3 400 0 0 0 20 15 36.700 0.520 0.320 8.700 0.000 0.001 0.000 0.000 0.059 N/A

4* 0 0 0 0 20 5 13.950 0.690 0.310 8.100 0.000 0.002 0.000 0.000 N/A N/A

5 200 0 0 0 20 5 27.025 0.700 0.300 11.100 0.000 0.002 0.000 0.000 0.045 N/A

6 200 35 0 0 20 5 32.825 0.640 0.470 13.700 0.000 0.005 0.001 0.001 0.080 N/A

7 200 34 38.17 0 20 5 32.325 1.400 1.100 14.000 0.010 0.009 0.020 0.006 0.074 N/A

8 200 32 38.19 13456 20 5 28.750 1.500 1.100 13.300 0.000 0.008 0.020 0.005 0.064 0.030

9 400 0 0 0 20 5 42.900 0.570 0.310 13.600 0.000 0.020 0.000 0.000 0.063 N/A

10 400 34 0 0 20 5 54.6 0.62 0.49 19.6 0.001 0.008 0 0 0.12 N/A

11 400 36 78.29 0 20 5 51.125 2.100 1.400 18.700 0.020 0.015 0.070 0.010 0.110 N/A

12 400 35 79.33 26224 20 5 44.750 2.200 1.600 17.700 0.001 0.014 0.070 0.010 0.090 0.030

13 0 0 37.95 0 20 5 13.225 1.400 1.000 8.200 0.000 0.005 0.060 0.010 N/A N/A

14 0 0 78.1 0 20 5 13.375 2.000 1.300 8.500 0.000 0.009 0.150 0.020 N/A N/A

* Denotes Baseline Load Test

How do we pull all this together?!Pharma Content Database Distribution

Substitute “F1” with SQL Server number to generate unique DB’sFarm 1: 2 SQLFarm 2: 1 SQL165 Content Databases!

Database # Content Database Name Database Size(TB)

# of Site Collections

Site Collection Names Subsites under each site collection

1 CDB_F1_350_1 0.35 30 Site_F1_350_1_1 thru 30 102 CDB_F1_350_2 0.35 60 Site_F1_350_2_1 thru 60 103 CDB_F1_350_3 0.35 60 Site_F1_350_3_1 thru 60 104 CDB_F1_350_4 0.35 60 Site_F1_350_4_1 thru 60 105 CDB_F1_350_5 0.35 60 Site_F1_350_5_1 thru 60 106 CDB_F1_350_6 0.35 60 Site_F1_350_6_1 thru 60 107 CDB_F1_350_7 0.35 60 Site_F1_350_7_1 thru 60 108 CDB_F1_300_1 0.3 30 Site_F1_300_1_1 thru 30 109 CDB_F1_300_2 0.3 60 Site_F1_300_2_1 thru 60 10

10 CDB_F1_300_3 0.3 60 Site_F1_300_3_1 thru 60 1011 CDB_F1_300_4 0.3 60 Site_F1_300_4_1 thru 60 1012 CDB_F1_300_5 0.3 60 Site_F1_300_5_1 thru 60 1013 CDB_F1_300_6 0.3 60 Site_F1_300_6_1 thru 60 1014 CDB_F1_300_7 0.3 60 Site_F1_300_7_1 thru 60 1015 CDB_F1_300_8 0.3 60 Site_F1_300_8_1 thru 60 1016 CDB_F1_250_1 0.25 30 Site_F1_250_1_1 thru 30 1017 CDB_F1_250_2 0.25 60 Site_F1_250_2_1 thru 60 1018 CDB_F1_250_3 0.25 60 Site_F1_250_3_1 thru 60 1019 CDB_F1_250_4 0.25 60 Site_F1_250_4_1 thru 60 1020 CDB_F1_250_5 0.25 60 Site_F1_250_5_1 thru 60 1021 CDB_F1_250_6 0.25 60 Site_F1_250_6_1 thru 60 1022 CDB_F1_250_7 0.25 60 Site_F1_250_7_1 thru 60 1023 CDB_F1_250_8 0.25 60 Site_F1_250_8_1 thru 60 1024 CDB_F1_250_9 0.25 60 Site_F1_250_9_1 thru 60 1025 CDB_F1_250_10 0.25 60 Site_F1_250_10_1 thru 60 1026 CDB_F1_250_11 0.25 60 Site_F1_250_11_1 thru 60 1027 CDB_F1_250_12 0.25 60 Site_F1_250_12_1 thru 60 1028 CDB_F1_250_13 0.25 60 Site_F1_250_13_1 thru 60 1029 CDB_F1_250_14 0.25 60 Site_F1_250_14_1 thru 60 1030 CDB_F1_250_15 0.25 60 Site_F1_250_15_1 thru 60 1031 CDB_F1_250_16 0.25 60 Site_F1_250_16_1 thru 60 1032 CDB_F1_250_17 0.25 60 Site_F1_250_17_1 thru 60 1033 CDB_F1_250_18 0.25 60 Site_F1_250_18_1 thru 60 1034 CDB_F1_200_1 0.2 30 Site_F1_200_1_1 thru 30 1035 CDB_F1_200_2 0.2 60 Site_F1_200_2_1 thru 60 1036 CDB_F1_200_3 0.2 60 Site_F1_200_3_1 thru 60 1037 CDB_F1_200_4 0.2 60 Site_F1_200_4_1 thru 60 1038 CDB_F1_200_5 0.2 60 Site_F1_200_5_1 thru 60 1039 CDB_F1_200_6 0.2 60 Site_F1_200_6_1 thru 60 1040 CDB_F1_200_7 0.2 60 Site_F1_200_7_1 thru 60 1041 CDB_F1_200_8 0.2 60 Site_F1_200_8_1 thru 60 1042 CDB_F1_200_9 0.2 60 Site_F1_200_9_1 thru 60 1043 CDB_F1_200_10 0.2 60 Site_F1_200_10_1 thru 60 1044 CDB_F1_200_11 0.2 60 Site_F1_200_11_1 thru 60 1045 CDB_F1_200_12 0.2 60 Site_F1_200_12_1 thru 60 1046 CDB_F1_200_13 0.2 60 Site_F1_200_13_1 thru 60 1047 CDB_F1_200_14 0.2 60 Site_F1_200_14_1 thru 60 1048 CDB_F1_200_15 0.2 60 Site_F1_200_15_1 thru 60 1049 CDB_F1_200_16 0.2 60 Site_F1_200_16_1 thru 60 1050 CDB_F1_200_17 0.2 60 Site_F1_200_17_1 thru 60 1051 CDB_F1_150_1 0.15 30 Site_F1_150_1_1 1052 CDB_F1_150_2 0.15 60 Site_F1_150_1_1 thru 60 1053 CDB_F1_150_3 0.15 60 Site_F1_150_2_1 thru 60 1054 CDB_F1_100_1 0.1 30 Site_F1_100_1_1 1055 CDB_F1_100_2 0.1 60 Site_F1_100_2_1 thru 60 10

Totals 13.4 3120

How do we pull all this together?!Pharma Data Load Statistics

Farm 1SQL 1-1 Total Docs 23,196,327.00 SQL 1-1 Total TB 13.16TB

SQL 1-2 Total Docs 23,110,478.00 SQL 1-2 Total TB 13.12TBTotal TB 26.29TB

Type Count %DOCX 23,393,946.00 50.52%XLSX 9,364,053.00 20.22%PPTX 7,007,626.00 15.13%HTML 2,334,473.00 5.04%TXT 2,391,297.00 5.16%PDF 1,815,385.00 3.92%Total Docs 46,306,780.00 100%

Total Load Time 10.92Days

Farm 2SQL 1-1 Total Docs 25,217,577.00 SQL 1-1 Total TB 13.10TB

Total TB 13.10TB

Type Count %DOCX 12,668,489.00 50.24%XLSX 5,071,017.00 20.11%PPTX 3,801,674.00 15.08%HTML 1,273,835.00 5.05%TXT 1,295,295.00 5.14%PDF 1,107,267.00 4.39%Total Docs 25,217,577.00 100%

Total Load Time 7.96Days

Architectural Design Statistical ResultsTesting Results – 300GB Content DatabasesRUN: ALL.300.501 ALL.300.501 ALL.300.501 ALL.300.501 ALL.300.501USERS: 200 500 1000 2000 5000

Avg. Response Time (Sec) 0.27 0.68 1.37 2.8 7.4Avg. First Byte Time (Sec) 0.27 0.65 1.26 2.5 5.7Avg. Page Time (Sec) 0.84 2.0 3.60 7.5 16.7Avg. Failed Requests 9 16 12 17 22Avg. Reqs/Sec 665 787 779 493 573Actual RPS * 298 352 349 221 256Avg. WFE CPU Util 53.28% 57.3% 58.18% 32.67% 40.6%Proc's 16(8x2) 16(8x2) 16(8x2) 16(8x2) 16(8x2)Memory 32 32 32 32 32Think Time 0 0 0 0 0

Architectural Design Statistical ResultsTesting Results – 350GB Content Databases

RUN:350.300.5

01350.300.5

01350.300.5

01350.300.5

01350.300.5

01350.300.5

01

USERS=: 200 500 1000 2000 3000 5000

Avg. Response Time (Sec) 0.34 0.51 1.17 2.4 2.5 5.04Avg. First Byte Time (Sec) 0.33 .50 1.15 2.3 2.9 4.06

Avg. Page Time (Sec) 1.02 1.50 3.59 7.4 8.6 12.2

Avg. Failed Requests 123 282 258 261 371 520

Avg. Reqs/Sec 537 913 795 617 773 842

Actual RPS * 241 224 356 277 347 378

Avg. WFE CPU Util 42.73% 71.4% 65.88% 46.55% 59.05% 72.77%

Proc's 16(8x2) 16(8x2) 16(8x2) 16(8x2) 16(8x2) 16(8x2)

Memory 32 32 32 32 32 32

Think Time 0 0 0 0 0 0

Architectural Design Statistical ResultsTesting Results – 250GB Content Databases

RUN:250.300.50

1250.300.50

1250.300.50

1250.300.50

1250.300.50

1250.300.50

1

USERS: 200 500 1000 2000 3000 5000

Avg. Response Time (Sec) 0.33 0.60 1.3 2.58 3.9 6.4Avg. First Byte Time (Sec) 0.33 0.59 1.2 2.17 3.2 5.0

Avg. Page Time (Sec) 1.0 1.81 3.7 6.6 9.8 14.6

Avg. Failed Requests 0 3 1 2 2 3

Avg. Reqs/Sec 554 764 769 714 652 648

Actual RPS * 248 343 345 320 292 291

Avg. WFE CPU Util 44.15% 59.35% 58.22% 51% 48.25% 46.6%

Proc's 16(8x2) 16(8x2) 16(8x2) 16(8x2) 16(8x2) 16(8x2)

Memory 32 32 32 32 32 32

Think Time 0 0 0 0 0 0

Architectural Design Statistical ResultsTesting Results – 150GB Content Databases

RUN:150.300.50

1150.300.50

1150.300.50

1150.300.50

1150.300.50

1150.300.50

1USERS=: 200 500 1000 2000 3000 5000

Avg. Response Time (Sec) 0.27 0.60 1.3 1.94 2.5 4.3Avg. First Byte Time (Sec) 0.27 0.59 1.2 1.84 2.4 3.7Avg. Page Time (Sec) 0.81 1.78 3.8 5.46 7.3 11.0Avg. Failed Requests 507 1079 875 939 1238 1616Avg. Reqs/Sec 663 927 737 789 706 767Acutal RPS * 297 416 330 354 317 344Avg. WFE CPU Util 53.55% 77.45% 77.95% 67.45% 57.8% 70.0%Proc's 16(8x2) 16(8x2) 16(8x2) 16(8x2) 16(8x2) 16(8x2)Memory 32 32 32 32 32 32Think Time 0 0 0 0 0 0

© 2009 Microsoft Corporation. All rights reserved. Microsoft, Windows, Windows Vista and other product names are or may be registered trademarks and/or trademarks in the U.S. and/or other countries.The information herein is for informational purposes only and represents the current view of Microsoft Corporation as of the date of this presentation. Because Microsoft must respond to changing market conditions, it should not be interpreted to be a commitment on the part of Microsoft, and Microsoft cannot guarantee the accuracy of any information provided after the date of this presentation. MICROSOFT MAKES NO WARRANTIES, EXPRESS,

IMPLIED OR STATUTORY, AS TO THE INFORMATION IN THIS PRESENTATION.