Best Practices in Virtualizing RDS and VDI: THE Virtual Reality Check Ruben Spruijt @rspruijt.

download Best Practices in Virtualizing RDS and VDI: THE Virtual Reality Check Ruben Spruijt @rspruijt.

If you can't read please download the document

Transcript of Best Practices in Virtualizing RDS and VDI: THE Virtual Reality Check Ruben Spruijt @rspruijt.

  • Slide 1
  • Best Practices in Virtualizing RDS and VDI: THE Virtual Reality Check Ruben Spruijt @rspruijt
  • Slide 2
  • Project: V irtual R eality C heck
  • Slide 3
  • Ruben Spruijt, PQR
  • Slide 4
  • A-Team Jeroen van de Kamp: [email protected] Ruben Spruijt: [email protected]
  • Slide 5
  • Introduction Phase 1 & 2 Phase 3 (VDI Deep Dive) Phase 4 (AppVirt) - PREVIEW Unpublished results Agenda.
  • Slide 6
  • .
  • Slide 7
  • www.VirtualRealityCheck.net.
  • Slide 8
  • Performance Analysis & Review VDI + SBC.
  • Slide 9
  • Independent & Unbiased.
  • Slide 10
  • Latest Gear.
  • Slide 11
  • .
  • Slide 12
  • Slide 13
  • Slide 14
  • Slide 15
  • Slide 16
  • Slide 17
  • Slide 18
  • Slide 19
  • SBC (TS/XenAPP) Terminal Server AD File Share Hypervisor Terminal Server Logging
  • Slide 20
  • VDI Hypervisor 7 7 AD File Share Logging 7 7 7 7 7 7
  • Slide 21
  • Office 2007: Outlook, Word, PowerPoint Excel PDF printer & Adobe PDF Internet Explorer (multiple sites + Flash Video) FreeMind VSI workload (14 minute loop)
  • Slide 22
  • Slide 23
  • Phase 1 & 2.
  • Slide 24
  • Virtualization TS/XenApp: Yes we can! Sweet spot = performance + licensing Best-Practices No vCPU overcommit + 2vCPU minimum Dedicated Host / Reservation, Enforce vMMU! (x86 - vSphere) Intel Nehalem/Westmere: impressive Hyper-Threading + EPT-D = Doubling Capacity Conclusions Phase 1 & 2.
  • Slide 25
  • Phase 3.
  • Slide 26
  • Test platform Phase 3: VMware vSphere 4.x Server Brand/ModelHPDL380G6 CPU2 x Intel Quad core [email protected] Nehalem (16 logical cpu!) Memory96GB DDR3 Disk8 x 146Gb, 820.2Gb, dual port 10.000RPM Serial SCSI RAID levelRAID-5 with online spare RAID controllerHP Smart Array P400i, with 512MB and Battery Backed Write Cache NICNC373i Gigabit Adapters, Broadcom 5708.
  • Slide 27
  • VSI vs VRC Optimizations.
  • Slide 28
  • VRC optimizations: Standard VSI Tuning.
  • Slide 29
  • VRC optimizations: Extra Disabled Services.
  • Slide 30
  • VRC optimizations: Scheduled Tasks.
  • Slide 31
  • Project VRC must emphasize that it is crucial to test and validate these optimizations in your own VDI deployment
  • Slide 32
  • VDI + Storage.
  • Slide 33
  • Windows XP Windows 7 + VSI Optimizations Windows 7 + VRC Optimizations Understanding I/O
  • Slide 34
  • Boot and Logon First VSI loop Second VSI loop Idle + 6 Apps Idle + No Apps Understanding I/O
  • Slide 35
  • Boot and Logon First VSI loop Second VSI loop Idle + 6 Apps Idle + No Apps Understanding I/O
  • Slide 36
  • Boot and Logon
  • Slide 37
  • Boot and Logon; normalized
  • Slide 38
  • Boot and Logon, Read vs Write I/O
  • Slide 39
  • Boot and Logon First VSI loop Second VSI loop Idle + 6 Apps Idle + No Apps Understanding I/O
  • Slide 40
  • First VSI Loop
  • Slide 41
  • First VSI Loop: normalized
  • Slide 42
  • First VSI Loop, Read vs Write I/O
  • Slide 43
  • Boot and Logon First VSI loop Second VSI loop Idle + 6 Apps Idle + No Apps Understanding I/O.
  • Slide 44
  • Second VSI Loop.
  • Slide 45
  • Second VSI Loop: normalized.
  • Slide 46
  • Second VSI Loop: Read vs Write I/O.
  • Slide 47
  • Boot and Logon First VSI loop Second VSI loop Idle + 6 Apps Idle + No Apps Understanding I/O.
  • Slide 48
  • Idle I/O With 6 Applications Running.
  • Slide 49
  • Idle I/O With 6 Applications Running: Normalized.
  • Slide 50
  • Boot and Logon First VSI loop Second VSI loop Idle + 6 Apps Idle + No Apps Understanding I/O.
  • Slide 51
  • Idle I/O With No Running Apps.
  • Slide 52
  • Idle I/O With No Running Apps: Normalized.
  • Slide 53
  • 1GB vs 2GB Memory (VRC Opt).
  • Slide 54
  • .
  • Slide 55
  • Fixed Or Automatic Page File Size (VRC Opt).
  • Slide 56
  • IO impact Avirus on VDI
  • Slide 57
  • VIRUSSCANNER: TOTAL I/Os
  • Slide 58
  • To be continued..
  • Slide 59
  • Conclusion I/Os.
  • Slide 60
  • Windows 7 more than Windows XP Boot First Loop Windows 7 less than Windows XP Second Loop Idle Best practice: Use VRC Optimizations fix page file A/V Conclusion I/Os
  • Slide 61
  • Red Bull
  • Slide 62
  • VMware vSphere: HaltingIdleMsecPenalty = HIMP Default = 100 Red Bull
  • Slide 63
  • VDI on Red Bull; Windows XP
  • Slide 64
  • Slide 65
  • VDI on Red Bull; Windows 7
  • Slide 66
  • Slide 67
  • Hyper-Threading Disabled
  • Slide 68
  • Slide 69
  • HIMP = VMware vSphere; >> Hyper-V by design fast #LOL 23% Improvement Windows XP 22% Improvement Windows 7 Only on Hyper-Threading HaltingIdleMsecPenaltyMax HIMP Conclussion.
  • Slide 70
  • Phase 3 - VSI Results.
  • Slide 71
  • Slide 72
  • VSI vs VRC Optimizations.
  • Slide 73
  • 1 vCPU vs 2 vCPU.
  • Slide 74
  • Memory Overcommit (vSphere) Win7 120VM Pre-Booted (96GB RAM).
  • Slide 75
  • SBC vs VDI
  • Slide 76
  • Slide 77
  • Slide 78
  • VSI vs VRC Optimizations 1 vs 2vCPU Memory overcommit SBC vs VDI VSImax Conclusions
  • Slide 79
  • Preview Phase 4 PREVIEW.
  • Slide 80
  • A-Team
  • Slide 81
  • VSIMax Local vs Streaming I/O Impact Phase 4: Impact AppVirt on VDI
  • Slide 82
  • 3 major AppVirt vendors: Citrix Application streaming Microsoft App-V VMware ThinApp Office 2007 suite virtualized as 1 package Different scenarios: streamed, precached, shared cache AppVirt on VDI: test setup
  • Slide 83
  • Project VRCs goal is to investigate overall performance impact of AppVirt in VDI. Project VRC does not recommend virtualizing the Microsoft Office suite as an overall best practice.
  • Slide 84
  • Maximum users on a host VSIMax results
  • Slide 85
  • App-V vs. Local installed %
  • Slide 86
  • Impact VSIMax: ~24% Different configurations doesnt impact VSIMax App-V conclusions
  • Slide 87
  • streamed vs. Local installed % Bug > Fix!
  • Slide 88
  • Pre-cached vs. Local installed % Bug > Fix!
  • Slide 89
  • VDI + Storage - IOPS.
  • Slide 90
  • IO reads: stream vs. local %
  • Slide 91
  • IO reads: cached vs. local %
  • Slide 92
  • IO writes: stream vs. local %
  • Slide 93
  • IO writes: pre-cached vs. local %
  • Slide 94
  • Worst/worse case scenario or reality?: Impact VSIMax: -24% Streaming apps = 22-45% less READ IOs Streaming apps = 20-45% more WRITE IOs Check response times in whitepaper (release soon) Application Virtualization IS key in Optimized (virtual) Desktop Appvirt: Conclusion
  • Slide 95
  • VDI + Storage - IOPS.
  • Slide 96
  • Fusion IO.
  • Slide 97
  • .
  • Slide 98
  • .
  • Slide 99
  • .
  • Slide 100
  • .
  • Slide 101
  • Finally..
  • Slide 102
  • Hyper-V SP1 Impact of Dynamic Memory in SP1 =
  • Slide 103
  • Windows 7 Dynamic Memory (SP1)
  • Slide 104
  • Hyper-V SP1 Impact of Dynamic Memory in SP1 =
  • Slide 105
  • To be continued..
  • Slide 106
  • Head-2-Head: VDI on Hyper-V/vSphere/Xenserver (Phase 5) AntiVirus impact and optimization in VDI (Phase 6) VDI: Remoting Protocols; impact on platform (Phase 7) VDI: Storage Optimization/Solutions (TBD).. Ideas!? >> @ VRC vNext 2011.
  • Slide 107
  • www.virtualrealitycheck.net www.twitter.com/ProjectVRC VDI + Storage = Deep Impact http://bit.ly/5HTajV (PQR,@brugh) Special thanks to: Sven Huisman (PQR, @svenh) Related Info
  • Slide 108
  • .
  • Slide 109
  • .
  • Slide 110
  • .
  • Slide 111
  • Speaker info: please do not delete the slides in this section Show these slides at the end of your session before going to Thank you page.
  • Slide 112
  • Stay up to date with TechNet Belux Register for our newsletters and stay up to date: http://www.technet-newsletters.be Technical updates Event announcements and registration Top downloads Join us on Facebook http://www.facebook.com/technetbe http://www.facebook.com/technetbelux LinkedIn: http://linkd.in/technetbelux/ Twitter: @technetbelux Download MSDN/TechNet Desktop Gadget http://bit.ly/msdntngadget
  • Slide 113
  • TechDays 2011 On-Demand Watch this session on-demand via TechNet Edge http://technet.microsoft.com/fr-be/edge/ http://technet.microsoft.com/nl-be/edge/ Download to your favorite MP3 or video player Get access to slides and recommended resources by the speakers
  • Slide 114
  • Ruben Spruijt: [email protected] Jeroen van de Kamp: [email protected]