University of
Michigan Administrative Information Services
Server Virtualization Technologies: Uses, Comparisons, and Implications
David Sweetman Windows Enterprise Systems Admin Administrative Information Services University of Michigan
dsweetma@[Link]
Presentation Overview
The What and Why of virtualization Comparing Product Features Comparing Product Performance Evaluating Physical Servers for virtualization Costs Questions
Spring, 2005
Windows Virtualization
What is server virtualization? Creating multiple logical server OS instances on one physical piece of hardware All HW drivers are virtualized same virtual HW regardless of physical HW Each virtual machine is completely independent of the others and doesnt realize its virtualized
Spring, 2005 Windows Virtualization 3
Why virtualize?
More efficient HW utilization More efficient staff Long-term matching resources & needs Quick and nimble server provisioning Testing & Troubleshooting More effective redundancy HW maintenance w/o app downtime Simplify system imaging Disaster Recovery
Spring, 2005 Windows Virtualization 4
HW Utilization Facts
Individual ebb and flow of resources Cumulative usage of 28 servers in the MAIS data center evaluated for virtualization: 44GB RAM, 138.15Ghz CPU, and 1323GB HD 45% of RAM not used 99.9% of time. 25% of RAM never used concurrently. 85% of CPU not used 99.9% of time. 81% of CPU never used concurrently. 68% of hard disk space unused
Spring, 2005 Windows Virtualization 5
Server Local Disk
More Efficient Hard SAN Manager Hard Disk Utilization 48 8 40 Disk Utilization IIS app test 68 9 59
68 34 34 68 17 136 34 17 68 68 68 34 68 170 34 13 7 24 31 9 56 6 6 16 11 10 13 6 88 7 55 27 10 37 8 80 28 11 52 57 58 21 62 82 27
Total (GB)
Used (GB)
Free (GB)
TNG Scheduling PeopleSoft 8 HE PeopelSoft 8 FIN IIS / SQL:Research app Small use Citrix File Servers Stat Version Control Stat Version Control SQL: eLearning dev IIS: eLearning dev SQL: eLearning Prod IIS: eLearning Prod Machine Room environ IIS document server Domain Controller
Total: 1323 GB Used: 418 GB Free: 905 GB (68% unused)
SAN in 30GB chunks 1 fibre channel >1 server Virtual HDs more granular Share free space allocate as needed Windows Virtualization
6
Spring, 2005
Virtualization vs. Consolidation Virtualized servers = separate OSes Consolidation = same OS Virtualized servers must each be administered, patched, etc. Consolidated applications can introduce conflicts and support issues
Spring, 2005
Windows Virtualization
Virtual Host Licensing Windows and other Microsoft per-server apps are licensed per virtual server. (1 physical server w/ 6 virtual Windows servers = 6-7 licenses needed) As of 4/1/2005, Microsoft per-processor licenses are per physical processor (1 physical server w/ 3 virtual SQL Servers sharing 1 CPU = 1 per-processor license) Virtualization savings are not in licenses. Check with other vendors.
Spring, 2005 Windows Virtualization 8
Virtualization Software MS Virtual PC 2004 workstation only VMWare Workstation 5 workstation only MS Virtual Server 2005, Standard (4p) MS Virtual Server 2005, Enterprise (32p) VMWare GSX Server 3.1 VMWare ESX Server 2.5
Spring, 2005 Windows Virtualization 9
Common Features
Up to 3.6GB RAM per virtual host Web-based console for administration Host OS sees HT CPU, virtual do not VMs consist of 1 config file & 1 file / HD VMs can mount physical CDs or ISOs VMs can be multi-homed Up to 64 VMs per host server Highly scriptable extensive API Granular permissions for individual VMs Detailed logging
Spring, 2005 Windows Virtualization 10
MS Virtual Server 2005 Targeted to increase efficiency in testing and development, and re-hosting Up to 1 processor per virtual host Windows = underlying host OS Only Windows VMs supported No USB support 2 processor SMP coming soon
Spring, 2005
Windows Virtualization
11
VMWare ESX Server 2.5 Targeted at mission-critical enterprise services Up to 2 processors per host Custom Linux = underlying OS Windows & Linux VMs supported Dedicated NIC for admin (2 total min) USB support 4 proc SMP coming soon
Spring, 2005 Windows Virtualization 12
Do I need to know Linux? VMWare ESX Server is based on Linux All administration is possible through web Dont need any Linux experience for installation or ongoing admin SSH and SFTP access to server Used?
Installed backup software sFTPed ISOs to server
Spring, 2005 Windows Virtualization 13
Managing Virtual Servers Web site is primary interface Attach to VM console
Virtual Server = ActiveX control VMWare = separate application
Reboot, power on, power off Create and manage VMs Allocate hardware resources Mount CDs and floppies View recent performance data
Spring, 2005 Windows Virtualization 14
VS Screenshot
Spring, 2005
Windows Virtualization
15
VMWare Screenshot
Spring, 2005
Windows Virtualization
16
Hyper-threading One physical CPU seen as 2 logical Both products see HT, non-HT VMs Slows virtualization performance 1 HT CPU < 2 Phy CPU 0-20% performance increase over no HT [Link] ead/
Spring, 2005
Windows Virtualization
17
RAM Allocation Virtual Server: Max <= total physical memory VMWare: Max <> total physical
RAM Ballooning RAM pooled across multiple VMs Enables more efficient RAM utilization If max out, goes to paging file
Windows Virtualization 18
Spring, 2005
VS Screenshot
Spring, 2005
Windows Virtualization
19
VMWare Screenshot
Spring, 2005
Windows Virtualization
20
Monitoring MOM (or other host monitoring): Monitors VMs like physical Virtual Server: MOM Management Pack
Integrates into MOM framework Monitor overall host and VM servers
VMWare: vmkusage VMWare: VirtualCenter
Database back-end across all servers
Spring, 2005 Windows Virtualization 21
Virtual Center Central monitoring and management in VMWare environment Manage all VMs from one interface Additional software / license Management application Set thresholds and actions like MOM SQL or Oracle DB backend Assign privileges via NTFS
Spring, 2005 Windows Virtualization 22
Virtual Center Screenshot
Spring, 2005
Windows Virtualization
23
Converting Physical Server
Both MS & VMWare offer tools to create virtual systems from physical Physical HW drivers replaced by VM Ideal for the truly unique server (highly customized) Both vendors recommend loading virtual servers from scratch Slow for both vendors 6h / 4GB image VSMT (Virtual Server Migration Tool)
many prereqs (DHCP, ADS, SQL) Not in one month eval
P2V (Physical 2 Virtual)
Simple boot CD and server piece Licensed per use
Spring, 2005 Windows Virtualization 24
VMotion Enables seamless transition of live virtual host between physical servers Dynamic Resource Allocation across servers respond to load changes HW maintenance
Spring, 2005
Windows Virtualization
25
Best Practices Plan out server allocations Create gold image base OS kept upto-date patches duplicate for new VMs Use ISOs for CD access Use standard backup and restore Take system images as needed
Spring, 2005
Windows Virtualization
26
Summary of VMWare differences More comprehensive web GUI (for example, deleting hosts & HDs) Support for dual processor virtuals Support for Linux virtuals Virtual Center: central management Easy-to-use physical-to-virtual support VMotion: seamlessly move virtual servers between physical hosts
Spring, 2005 Windows Virtualization 27
Testing Environment
One month each was spent evaluating MS Virtual Server & VMWare ESX Server Identical testing was attempted on each. Load and usability testing: Win 2000, 2003, IIS5, IIS6, SQL Server 2000, 3rd party apps Test hardware
1.4Ghz x 4 physical processors (8 w/ HT) 8GB of RAM 60GB fibre-channel connected SAN space
Spring, 2005 Windows Virtualization 28
Performance Comparisons
Automated load test of Aspen 2.5 dev environment (Win 2000/IIS5 & Win 2000/SQL 2000) Citrix / TS load test w/ Helpdesk IIS6-based memory, CPU, disk, and network I/O testing SQL Server add, update, and delete testing Load testing both as isolated server and with other virtual server processing Normal usage w/o issue in all cases
Spring, 2005 Windows Virtualization 29
Performance Comparisons
Windows 2003 IIS6 and SQL 2000 perf compare
Physical CPU Memory Disk&NIC I/O SQL 100% 100% 100% 100% MSVS 94% 91% 101% 57% VMWare 80% 91% 101% 87%
VMWare CPU : hyper-threaded related, ~93% w/o VS SQL : VS 2005 SP1 has performance enhancements
Spring, 2005
Windows Virtualization
30
Performance Comparisons
Previous stats were isolated tests VMs wont be alone on physical host How does system perform w/ other VMs running assorted, intensive tasks?
RAM Virtual Server 2005 -/+ <10% VMWare ESX Server Same CPU -/+ <10% Same Disk - <5% - <5% Network - <5% - <5%
Spring, 2005
Windows Virtualization
31
IIS/SQL Load Test Results Mercury LoadRunner scripted test Overall performance
100@30/min: VM = 60% 1000@12/min: VM = 99%
What made it slow?
CPU queuing Memory, HD, NetIO nearly identical
Spring, 2005
Windows Virtualization
32
Terminal Services / Citrix Load Test Results
Currently 14 servers, 4procs (8HT), 4GB RAM load balancing ~700 concurrent CPU and RAM intensive apps ~60 users max per physical server CPU = bottleneck (logon & BusObj) 1CPU = 7 users max ; 2 CPU = 12 max 100 v 1CPU or 58 v 2CPU to match 14 physicals Recommendation: 2 CPU & only for small use
Spring, 2005 Windows Virtualization 33
Business Objects WebI dev
Win 2000 / IIS5 / 2400MB RAM / 1.4Ghz x 2 (no HT)
24 x 7 sampling RAM (MB) Nic KB HD KB /sec /sec Proc % Usage RAM (MB) Hours of Op sampling Nic KB /sec HD KB /sec Proc % Usage
100%847 99.99%839 99.9%823 99%816 95%813 90%809 Av755 StDev47
61 45 19 17 4 2 2 4
4324 1548 487 76 65 64 55 86
112 55 52 3 3 2 2 3
847 839 821 816 814 812 759 44
61 47 32 17 11 2 2 4
4324 97 85 67 65 63 52 136
34 24 22 7 6 5 1 1
Virtualize? Yes. 900 / 1.4Ghz
Spring, 2005 Windows Virtualization 34
PSoft 8 Fin Crystal/nVision: Dev
Win 2000 / 2300MB RAM / 1.1Ghz x 2 (no HT)
24 x 7 sampling RAM (MB) Nic KB HD KB /sec /sec Proc % Usage RAM (MB) Hours of Op sampling Nic KB /sec HD KB /sec Proc % Usage
100%835 99.99%822 99.9%821 99%814 95%798 90%786 Av644 StDev91
5284 31 4 3 3 2 51
5038 4912 4802 910 61 205 834
106 89 82 67 10 2 4 8
835 829 822 819 809 799 662 106
5284 4542 72 5 3 3 3 87
2529 1555 1494 1015 100 32 49 181
106 92 75 13 3 2 2 3
2973 5005
Virtualize? Yes. 900 / 1.4Ghz Spring, 2005 Windows Virtualization 35
PSoft8 HE Crystal/nVision - Prod
Win 2000 / 1500MB RAM / 2.8Ghz x 1 (w/ HT)
24 x 7 sampling RAM (MB) Nic KB /sec HD KB Proc % /sec Usage RAM (MB) Hours of Op sampling Nic KB /sec HD KB /sec Proc % Usage
100%716 99.99%710 99.9%620 99%534 95%483 90%447 Av378 StDev67
11499 9803 1422 119 8 1 10 162
3421 350 3379 329 2440 244 2304 192 183 50 105 356 34 23 23 7
716 713 691 460 440 437 363 63
9437 3493 674 119 11 1 7 109
3421 3379 779 351 74 51 49 159
348 333 267 220 32 22 21 6
Virtualize? NOT at this time CPU needs too high Spring, 2005 Windows Virtualization 36
sumTotal Aspen 2.5 eLearning
Win 2000 / SQL 2000 / 2358MB RAM / 1.9Ghz x 2 (w/ HT)
24 x 7 sampling RAM (MB) Nic KB /sec HD KB /sec Proc % RAM Usage (MB) Hours of Op sampling Nic KB /sec HD KB /sec Proc % Usage
100%2077 99.99%2075 99.9%2073 99%1984 95%1777 90%1670 Av1628 StDev76
9061 5865 2667 91 68 5 16 173
4477 3682 3673 3626 839 517 236 505
277 233 216 138 101 41 24 5
2077 2075 2073 2064 1684 1665 1636 60
406 404 206 70 67 3 5 20
1047 1039 971 827 623 459 166 183
155 149 138 125 59 30 21 4
Virtualize? Yes 2300MB / 1.4Ghz x 2 Spring, 2005
Note: high NIC=sync ; CPU=imp/exp Windows Virtualization 37
Domain Controllers
Win 2003 / 2000MB RAM / 700Mhz x 4 (no HT)
24 x 7 sampling RAM (MB) Nic KB /sec HD KB /sec Proc % RAM Usage (MB) Hours of Op sampling Nic KB /sec HD KB /sec Proc % Usage
100%776 99.99%771 99.9%768 99%753 95%713 90%707 Av633 StDev74
5677 5326 2131 51 24 15 12 138
4298 3674 3440 1972 140 91 128 302
146 131 78 43 12 10 8 3
767 766 757 753 713 707 646 56
457 98 93 42 27 20 7 11
1237 195 194 180 90 88 78 41
92 51 16 13 11 10 8 1
Virtualize? Yes 850MB / 1.4Ghz Spring, 2005 Windows Virtualization 38
Univ of Michigan - Flint VMWare ESX Server Determining factor: Linux support & MS Virtual Server wasnt available Several years of experience, starting with GSX, public web services, online teaching, real video server, internal file/print, 46v on 5 physical (15 on 1), <10% slower, Dell 2650s & 4600s, 2 proc, 12GB RAM
Spring, 2005 Windows Virtualization 39
NC State University MS Virtual Server 2005 Determining factor: Cost PeopleSoft v8 Crystal/nVision app servers: 18 virtual servers, 7 physical servers, dual Xeon >2GB, physical v. virtual head-to-head, little difference in performance.
Spring, 2005
Windows Virtualization
40
Potential Uses from Previous Presentations
NAP - Remediation Servers Big Red Button for critical fix assign additional resources Keynote - Reliability one of pillars of Trustworthy Computing Boston U Matt - NetReg peak usage first couple weeks of semester WSUS 3Ghz, 1GB RAM recommended sitting idle most of time? Decrease dev system allocation in busy times
Spring, 2005 Windows Virtualization 41
Pricing
MS Virtual Server 2005 (4CPU Server, 8GB RAM)
Win 2003 Std: up to 4 processors, Ent: up to 32 VS Std: 4proc/4GB; Ent: 8proc/32GB 2003 Ent/Std: ~$500+~$500 = ~$1000
VMWare Server ESX (4CPU other pricing scales)
ESX: $4500/phy server + $945/yr support ESX+SMP+V-agents: $6000/phy server + $1764/yr support
VMWare Add-ons
VirtualCenter server: $3000 + $1050/yr P2V Starter kit (25): $2000 + $420/yr
Spring, 2005 Windows Virtualization 42
Cost / Benefit Example VMWare Server ESX
$45K separate HW purchase price $29K + $2K/yr (ESX w/SMP): ~35%
MS Virtual Server Std
$33K separate HW purchase price $30K virtual HW + software: ~10%
Note: In both cases, estimates are conservative
Spring, 2005
Windows Virtualization
43
Summary / take-aways
More effective resource utilization and response to changing needs (5-15% to 60-70%) Virtual Server & VMWare = comparable performance, VMWare more isolated VMWare more feature-rich: SMP, VMotion, manage multiple servers VMWare costs more, but you can do more, virtualize more costly servers Both platforms have limits, active improvement
Spring, 2005
Windows Virtualization
44
Other Resources VMWare: [Link] Virtual Server: [Link]/virtualserver/ Rapid App: [Link]
Spring, 2005
Windows Virtualization
45
David Sweetman University of Michigan dsweetma@[Link]
Questions?
Spring, 2005
Windows Virtualization
46