SlideShare a Scribd company logo
1 of 34
Download to read offline
Increasing XenServer’s VM density
Jonathan Davies,
XenServer System Performance Lead
XenServer Engineering, Citrix
Cambridge, UK

24 Oct 2013

Jonathan Davies (Citrix)

Increasing XenServer’s VM density

24 Oct 2013

1 / 34
Outline

1

Scalability expectations

2

Hard limits

3

Soft limits

4

Benchmarks

Jonathan Davies (Citrix)

Increasing XenServer’s VM density

24 Oct 2013

2 / 34
Scalability expectations

Outline

1

Scalability expectations

2

Hard limits

3

Soft limits

4

Benchmarks

Jonathan Davies (Citrix)

Increasing XenServer’s VM density

24 Oct 2013

3 / 34
Scalability expectations

Users expect VM density to scale with hardware

Jonathan Davies (Citrix)

Increasing XenServer’s VM density

24 Oct 2013

4 / 34
Scalability expectations

Users expect VM density to scale with hardware

Jonathan Davies (Citrix)

Increasing XenServer’s VM density

24 Oct 2013

5 / 34
Scalability expectations

Users get upset when. . .

Jonathan Davies (Citrix)

Increasing XenServer’s VM density

24 Oct 2013

6 / 34
Scalability expectations

XenServer’s VM density scalability
hard density
limit :-(

hardware's
theoretical capacity

XS 6.1
(and earlier)

Jonathan Davies (Citrix)

Increasing XenServer’s VM density

24 Oct 2013

7 / 34
Scalability expectations

XenServer’s VM density scalability
hard density
limit :-(

hardware's
theoretical capacity

XS 6.1
(and earlier)

XS 6.2

practical density limit
(depending on nature of VMs)

Jonathan Davies (Citrix)

Increasing XenServer’s VM density

hard density
limit :-)

24 Oct 2013

8 / 34
Hard limits

Outline

1

Scalability expectations

2

Hard limits

3

Soft limits

4

Benchmarks

Jonathan Davies (Citrix)

Increasing XenServer’s VM density

24 Oct 2013

9 / 34
Hard limits

Enumerated causes of limitations

Hard limit 1: dom0 event channels
Cause of limitation XenServer uses a 32-bit dom0
This means 1,024 dom0 event channels
#define MAX_EVTCHNS(d) 
(BITS_PER_EVTCHN_WORD(d) *
BITS_PER_EVTCHN_WORD(d))

Various VM functions use a dom0 event channel
VM density hard limit 225 VMs per host (PV with 1 vCPU, 1 VIF, 1 VBD)
150 VMs per host (HVM with 1 vCPU, 1 VIF, 3 VBDs)
Mitigation for XS 6.2 Hack for dom0 to enjoy 4,096 event channels
→ 800 VMs per host (PV with 1 vCPU, 1 VIF, 1 VBD)
→ 570 VMs per host (HVM with 1 vCPU, 1 VIF, 3 VBDs)
Mitigation for future Change the ABI to provide unlimited event channels
→ this would remove the limit
Jonathan Davies (Citrix)

Increasing XenServer’s VM density

24 Oct 2013

10 / 34
Hard limits

Enumerated causes of limitations

Hard limit 2: blktap2 device minor numbers

Cause of limitation blktap2 only supports up to 1,024 minor numbers
(despite the kernel allowing up to 1,048,576)
#define MAX_BLKTAP_DEVICE

1024

Each virtual block device requires one device
VM density hard limit 341 VMs per host (with 3 disks per VM)
Mitigation for XS 6.2 Double this constant to 2,048
→ 682 VMs per host (with 3 disks per VM)
Mitigation for future Move away from blktap2 altogether?
→ this would remove the limit

Jonathan Davies (Citrix)

Increasing XenServer’s VM density

24 Oct 2013

11 / 34
Hard limits

Enumerated causes of limitations

Hard limit 3: number of aio requests

Cause of limitation Each blktap2 instance creates an asynchronous I/O
context for receiving 402 events.
Default system-wide number of aio requests was
444,416 in XS 6.1.
VM density hard limit 368 VMs per host (with 3 disks per VM)
Mitigation for XS 6.2 Set fs.aio-max-nr to 1,048,576
→ 869 VMs per host (with 3 disks per VM)
Mitigation for future Increase fs.aio-max-nr further
or use storage driver domains
→ this would remove the limit

Jonathan Davies (Citrix)

Increasing XenServer’s VM density

24 Oct 2013

12 / 34
Hard limits

Enumerated causes of limitations

Hard limit 4: dom0 grant references

Cause of limitation Windows VMs use receive-side copy (RSC) by default in XS 6.1.
netback allocates (at least) 22 grant-table entries per
virtual interface for RSC.
dom0 had a total of 8,192 grant-table entries in
XS 6.1.
VM density hard limit 372 VMs per host (with 1 interface per VM)
Mitigation for XS 6.2 Don’t use RSC in Windows VMs anyway
→ this removes the limit

Jonathan Davies (Citrix)

Increasing XenServer’s VM density

24 Oct 2013

13 / 34
Hard limits

Enumerated causes of limitations

Hard limit 5: connections to xenstored

Cause of limitation xenstored uses select(2), which can only listen on
1,024 file descriptors.
#define __FD_SETSIZE

1024

qemu opens 3 file descriptors to xenstored.
VM density hard limit 333 VMs per host (HVM)
Mitigation for XS 6.2 Make two qemu watches share a connection
→ 500 VMs per host (HVM)
Mitigation for future Upstream qemu doesn’t connect to xenstored
→ this will remove the limit

Jonathan Davies (Citrix)

Increasing XenServer’s VM density

24 Oct 2013

14 / 34
Hard limits

Enumerated causes of limitations

Hard limit 6: connections to consoled

Cause of limitation Similarly, consoled uses select(2)
Each PV domain opens 3 fds to consoled
VM density hard limit 341 VMs per host (PV)
Mitigation for XS 6.2 Use poll(2) rather than select(2) in consoled
→ this removes the limit

Jonathan Davies (Citrix)

Increasing XenServer’s VM density

24 Oct 2013

15 / 34
Hard limits

Enumerated causes of limitations

Hard limit 7: dom0 low memory

Cause of limitation Each running VM eats about 1MB of dom0 lowmem
VM density hard limit around 650 VMs per host
Mitigation for future Use a 64-bit dom0
→ this will remove the limit

Jonathan Davies (Citrix)

Increasing XenServer’s VM density

24 Oct 2013

16 / 34
Hard limits

Summary of hard limits

Summary of hard limits

Limits on number of HVM guests with 1 vCPU, 1 VBD, 1 VIF (with PV drivers)

Limitation
dom0 event channels
blktap minor numbers
aio requests
dom0 grant references
xenstored connections
consoled connections
dom0 low memory
Overall limit
Limited by

Jonathan Davies (Citrix)

XS 6.1
225
1024
1105
372
333
no limit
650
225
event channels

XS 6.2
800
2048
2608
no limit
500
no limit
650
500
xenstored

Increasing XenServer’s VM density

Future
no limit
no limit
no limit
no limit
no limit
no limit
no limit
very high
something else!

24 Oct 2013

17 / 34
Hard limits

Summary of hard limits

Summary of hard limits

Limits on number of HVM guests with 1 vCPU, 3 VBDs, 1 VIF (with PV drivers)

Limitation
dom0 event channels
blktap minor numbers
aio requests
dom0 grant references
xenstored connections
consoled connections
dom0 low memory
Overall limit
Limited by

Jonathan Davies (Citrix)

XS 6.1
150
341
368
372
333
no limit
650
150
event channels

XS 6.2
570
682
869
no limit
500
no limit
650
500
xenstored

Increasing XenServer’s VM density

Future
no limit
no limit
no limit
no limit
no limit
no limit
no limit
very high
something else!

24 Oct 2013

18 / 34
Hard limits

Summary of hard limits

Summary of hard limits

Limits on number of PV guests with 1 vCPU, 1 VBD, 1 VIF

Limitation
dom0 event channels
blktap minor numbers
aio requests
dom0 grant references
xenstored connections
consoled connections
dom0 low memory
Overall limit
Limited by

Jonathan Davies (Citrix)

XS 6.1
225
1024
368
no limit
no limit
341
650
225
event channels

XS 6.2
1000
2048
869
no limit
no limit
no limit
650
650
dom0 lowmem

Increasing XenServer’s VM density

Future
no limit
no limit
no limit
no limit
no limit
no limit
no limit
very high
something else!

24 Oct 2013

19 / 34
Hard limits

Summary of hard limits

500 Windows VMs on a host

Jonathan Davies (Citrix)

Increasing XenServer’s VM density

24 Oct 2013

20 / 34
Soft limits

Outline

1

Scalability expectations

2

Hard limits

3

Soft limits

4

Benchmarks

Jonathan Davies (Citrix)

Increasing XenServer’s VM density

24 Oct 2013

21 / 34
Soft limits

xenstored

High dom0 CPU utilisation by xenstored

top - 16:29:33 up 36 min, 1 user, load average: 0.80, 0.56, 0.47
Tasks: 132 total,
1 running, 131 sleeping,
0 stopped,
0 zombie
Cpu(s): 40.1%us, 40.0%sy, 0.0%ni, 17.6%id, 0.0%wa, 0.0%hi, 0.0%si, 0.0%st
Mem:
4186504k total,
443480k used, 3743024k free,
23696k buffers
Swap:
524280k total,
0k used,
524280k free,
132504k cached
PID
7339
6665
7225
7269
7413
7283
10938
16403
1
2
3

USER
root
root
root
root
root
root
root
root
root
root
root

PR
20
20
20
20
20
20
20
20
20
20
RT

Jonathan Davies (Citrix)

NI VIRT RES SHR S %CPU %MEM
0 6732 2240 840 S 80.2 0.1
0 4344 2636 584 S 0.4 0.1
0 48892 5356 1736 S 0.3 0.1
0 23704 3684 1308 S 0.3 0.1
0 195m 21m 8932 S 0.3 0.5
0 7492 4860 1200 S 0.3 0.1
0 29808 1856 956 S 0.3 0.0
0 2428 1104 824 R 0.3 0.0
0 2164 656 564 S 0.0 0.0
0
0
0
0 S 0.0 0.0
0
0
0
0 S 0.0 0.0

Increasing XenServer’s VM density

TIME+
0:10.22
0:04.03
0:03.35
0:03.47
0:10.28
0:08.65
0:00.40
0:02.31
0:00.83
0:00.00
0:00.01

COMMAND
xenstored
fe
xcp-rrdd
xcp-rrdd-iostat
xapi
xcp-rrdd-xenpm
v6d
top
init
kthreadd
migration/0

24 Oct 2013

22 / 34
Soft limits

xenstored

High dom0 CPU utilisation by xenstored

dom0 vCPUs

domU vCPUs

Jonathan Davies (Citrix)

Increasing XenServer’s VM density

24 Oct 2013

23 / 34
Soft limits

xenstored

High dom0 CPU utilisation by xenstored

xenstored's dom0 vCPU
other dom0 vCPUs

domU vCPUs

Jonathan Davies (Citrix)

Increasing XenServer’s VM density

24 Oct 2013

24 / 34
Soft limits

xenstored

High dom0 CPU utilisation by xenstored

Cause of limitation xenstored CPU utilisation bottleneck
Mitigation for XS 6.2 Reduce xenstore use by XenServer’s toolstack:
remove some spurious writes
replace polling with watching

Jonathan Davies (Citrix)

Increasing XenServer’s VM density

24 Oct 2013

25 / 34
Soft limits

qemu

High dom0 CPU utilisation due to qemu
top - 16:40:27 up 2:07, 1 user, load average: 89.62, 87.22, 76.90
Tasks: 1015 total, 65 running, 950 sleeping,
0 stopped,
0 zombie
Cpu(s): 23.4%us, 55.5%sy, 0.0%ni, 4.8%id, 0.0%wa, 0.0%hi, 15.4%si, 0.5%st
Mem:
4180480k total, 1615840k used, 2564640k free,
3804k buffers
Swap:
524280k total,
0k used,
524280k free,
122852k cached
PID
7143
6653
16496
16970
997
3195
3497
3562
3993
7597
8150
8679
8974
11937
12545
14053
17752

USER
root
root
tcpdump
root
65583
65684
65656
65685
65546
65659
65550
65627
65661
root
65556
65598
65567

PR NI VIRT RES SHR S
20
0
0
0
0 R
10 -10 12264 7796 1152 R
20
0 5508 2132 1248 R
20
0 2952 1552 736 R
20
0 24696 4732 1572 S
20
0 24632 4736 1572 S
20
0 24760 4740 1576 R
20
0 24696 4732 1572 S
20
0 24888 4744 1580 S
20
0 24632 4736 1576 S
20
0 24760 4740 1580 R
20
0 24632 4740 1576 R
20
0 24568 4736 1572 S
20
0
0
0
0 S
20
0 24824 4748 1584 S
20
0 24760 4736 1576 S
20
0 24952 4740 1576 S

Jonathan Davies (Citrix)

%CPU %MEM
33.9 0.0
31.8 0.2
10.5 0.1
6.3 0.0
3.1 0.1
3.1 0.1
3.1 0.1
3.1 0.1
3.1 0.1
3.1 0.1
3.1 0.1
3.1 0.1
3.1 0.1
3.1 0.0
3.1 0.1
3.1 0.1
3.1 0.1

Increasing XenServer’s VM density

TIME+
17:21.63
36:14.34
5:35.12
0:00.11
0:56.30
0:27.34
0:28.65
0:26.97
0:53.19
0:28.86
0:51.71
0:31.18
0:27.97
1:12.92
0:58.46
0:31.33
0:56.82

COMMAND
rpciod/0
ovs-vswitchd
tcpdump
top
qemu-dm
qemu-dm
qemu-dm
qemu-dm
qemu-dm
qemu-dm
qemu-dm
qemu-dm
qemu-dm
nfsiod
qemu-dm
qemu-dm
qemu-dm
24 Oct 2013

26 / 34
Soft limits

qemu

qemu burning dom0 CPU

200 idle Windows guests,
each qemu utilising 3% of a CPU
means

6 dom0 vCPUs wasted!

Jonathan Davies (Citrix)

Increasing XenServer’s VM density

24 Oct 2013

27 / 34
Soft limits

qemu

What is qemu busy doing?

Emulated device
USB
CD-ROM
Buffered I/O & RTC timer
Parallel port
Serial port
VNC
qemu monitor

qemu events per VM per second
221
38
13
1
1
1
1

Mitigation for XS 6.2 Use an event-channel for buffered I/O notifications
Provide options to disable all emulated devices

Jonathan Davies (Citrix)

Increasing XenServer’s VM density

24 Oct 2013

28 / 34
Benchmarks

Outline

1

Scalability expectations

2

Hard limits

3

Soft limits

4

Benchmarks

Jonathan Davies (Citrix)

Increasing XenServer’s VM density

24 Oct 2013

29 / 34
Benchmarks

Bootstorm

Booting 90 Win7 VMs
Time to fully boot 90 VMs (25 at a time)
2500
Tampa (XS 6.1)
Clearwater (XS 6.2)

Elapsed time (s)

2000

1500

1000

XS 6.2 is
60% faster

500

0
0

25

50

75

100

125

150

175

200

VM index
Machine used: Dell PowerEdge R815 (Quad 16-core AMD Opteron 6272 @ 2.1GHz)
Jonathan Davies (Citrix)

Increasing XenServer’s VM density

24 Oct 2013

30 / 34
Benchmarks

Bootstorm

Booting 120 Win7 VMs
Time to fully boot 120 VMs (25 at a time)
2500
Tampa (XS 6.1)
Clearwater (XS 6.2)

Elapsed time (s)

2000

1500

1000
XS 6.2 is
75% faster

500

0
0

25

50

75

100

125

150

175

200

VM index
Machine used: Dell PowerEdge R815 (Quad 16-core AMD Opteron 6272 @ 2.1GHz)
Jonathan Davies (Citrix)

Increasing XenServer’s VM density

24 Oct 2013

31 / 34
Benchmarks

Bootstorm

Booting 200 Win7 VMs
Time to fully boot 200 VMs (25 at a time)
2500
Tampa (XS 6.1)
Clearwater (XS 6.2)

XS 6.1 can't
even get 200
VMs running!

Elapsed time (s)

2000

1500
It took XS 6.2 just
13 minutes to boot
200 VMs
(on this hardware)

1000

500

0
0

25

50

75

100

125

150

175

200

VM index
Machine used: Dell PowerEdge R815 (Quad 16-core AMD Opteron 6272 @ 2.1GHz)
Jonathan Davies (Citrix)

Increasing XenServer’s VM density

24 Oct 2013

32 / 34
Benchmarks

LoginVSI

number of VMs performing acceptably

LoginVSI: number of usable Windows VMs

XS 6.2

XS 6.1

number of VMs running
Machine used: Quad 10-core Intel E7-4860 @ 2.27GHz
Jonathan Davies (Citrix)

Increasing XenServer’s VM density

24 Oct 2013

33 / 34
Questions

Questions

?

Jonathan Davies (Citrix)

Increasing XenServer’s VM density

24 Oct 2013

34 / 34

More Related Content

What's hot

Red Hat Enterprise Linux OpenStack Platform on Inktank Ceph Enterprise
Red Hat Enterprise Linux OpenStack Platform on Inktank Ceph EnterpriseRed Hat Enterprise Linux OpenStack Platform on Inktank Ceph Enterprise
Red Hat Enterprise Linux OpenStack Platform on Inktank Ceph EnterpriseRed_Hat_Storage
 
了解IO设备
了解IO设备了解IO设备
了解IO设备Feng Yu
 
XPDS16: Scope and Performance of Credit-2 Scheduler. - Anshul Makkar, Ctirix...
XPDS16:  Scope and Performance of Credit-2 Scheduler. - Anshul Makkar, Ctirix...XPDS16:  Scope and Performance of Credit-2 Scheduler. - Anshul Makkar, Ctirix...
XPDS16: Scope and Performance of Credit-2 Scheduler. - Anshul Makkar, Ctirix...The Linux Foundation
 
Disk reports predicted failure event
Disk reports predicted failure eventDisk reports predicted failure event
Disk reports predicted failure eventAshwin Pawar
 
DataStax NYC Java Meetup: Cassandra with Java
DataStax NYC Java Meetup: Cassandra with JavaDataStax NYC Java Meetup: Cassandra with Java
DataStax NYC Java Meetup: Cassandra with Javacarolinedatastax
 
On-Demand Image Resizing
On-Demand Image ResizingOn-Demand Image Resizing
On-Demand Image ResizingJonathan Lee
 
Memory, Big Data, NoSQL and Virtualization
Memory, Big Data, NoSQL and VirtualizationMemory, Big Data, NoSQL and Virtualization
Memory, Big Data, NoSQL and VirtualizationBigstep
 
GemStone/S Update
GemStone/S UpdateGemStone/S Update
GemStone/S UpdateESUG
 
Cassandra Performance Benchmark
Cassandra Performance BenchmarkCassandra Performance Benchmark
Cassandra Performance BenchmarkBigstep
 
CephFS in Jewel: Stable at Last
CephFS in Jewel: Stable at LastCephFS in Jewel: Stable at Last
CephFS in Jewel: Stable at LastCeph Community
 
Hadoop: the Big Answer to the Big Question of the Big Data
Hadoop: the Big Answer to the Big Question of the Big DataHadoop: the Big Answer to the Big Question of the Big Data
Hadoop: the Big Answer to the Big Question of the Big DataVictor Haydin
 
Scaling Cassandra for Big Data
Scaling Cassandra for Big DataScaling Cassandra for Big Data
Scaling Cassandra for Big DataDataStax Academy
 
Openv switchの使い方とか
Openv switchの使い方とかOpenv switchの使い方とか
Openv switchの使い方とかkotto_hihihi
 
Varnish in action pbc10
Varnish in action pbc10Varnish in action pbc10
Varnish in action pbc10Combell NV
 
Network in OpenStack: changes since Cactus and CloudPipe HA
Network in OpenStack: changes since Cactus and CloudPipe HANetwork in OpenStack: changes since Cactus and CloudPipe HA
Network in OpenStack: changes since Cactus and CloudPipe HAMirantis
 
Thinking outside the box, learning a little about a lot
Thinking outside the box, learning a little about a lotThinking outside the box, learning a little about a lot
Thinking outside the box, learning a little about a lotMark Broadbent
 
How to operate containerized OpenStack
How to operate containerized OpenStackHow to operate containerized OpenStack
How to operate containerized OpenStackNalee Jang
 

What's hot (20)

Red Hat Enterprise Linux OpenStack Platform on Inktank Ceph Enterprise
Red Hat Enterprise Linux OpenStack Platform on Inktank Ceph EnterpriseRed Hat Enterprise Linux OpenStack Platform on Inktank Ceph Enterprise
Red Hat Enterprise Linux OpenStack Platform on Inktank Ceph Enterprise
 
了解IO设备
了解IO设备了解IO设备
了解IO设备
 
XPDS16: Scope and Performance of Credit-2 Scheduler. - Anshul Makkar, Ctirix...
XPDS16:  Scope and Performance of Credit-2 Scheduler. - Anshul Makkar, Ctirix...XPDS16:  Scope and Performance of Credit-2 Scheduler. - Anshul Makkar, Ctirix...
XPDS16: Scope and Performance of Credit-2 Scheduler. - Anshul Makkar, Ctirix...
 
Ha opensuse
Ha opensuseHa opensuse
Ha opensuse
 
Disk reports predicted failure event
Disk reports predicted failure eventDisk reports predicted failure event
Disk reports predicted failure event
 
DataStax NYC Java Meetup: Cassandra with Java
DataStax NYC Java Meetup: Cassandra with JavaDataStax NYC Java Meetup: Cassandra with Java
DataStax NYC Java Meetup: Cassandra with Java
 
On-Demand Image Resizing
On-Demand Image ResizingOn-Demand Image Resizing
On-Demand Image Resizing
 
Memory, Big Data, NoSQL and Virtualization
Memory, Big Data, NoSQL and VirtualizationMemory, Big Data, NoSQL and Virtualization
Memory, Big Data, NoSQL and Virtualization
 
GemStone/S Update
GemStone/S UpdateGemStone/S Update
GemStone/S Update
 
Cassandra Performance Benchmark
Cassandra Performance BenchmarkCassandra Performance Benchmark
Cassandra Performance Benchmark
 
CephFS in Jewel: Stable at Last
CephFS in Jewel: Stable at LastCephFS in Jewel: Stable at Last
CephFS in Jewel: Stable at Last
 
Hadoop: the Big Answer to the Big Question of the Big Data
Hadoop: the Big Answer to the Big Question of the Big DataHadoop: the Big Answer to the Big Question of the Big Data
Hadoop: the Big Answer to the Big Question of the Big Data
 
Scaling Cassandra for Big Data
Scaling Cassandra for Big DataScaling Cassandra for Big Data
Scaling Cassandra for Big Data
 
Openv switchの使い方とか
Openv switchの使い方とかOpenv switchの使い方とか
Openv switchの使い方とか
 
Stabilizing Ceph
Stabilizing CephStabilizing Ceph
Stabilizing Ceph
 
Varnish in action pbc10
Varnish in action pbc10Varnish in action pbc10
Varnish in action pbc10
 
Network in OpenStack: changes since Cactus and CloudPipe HA
Network in OpenStack: changes since Cactus and CloudPipe HANetwork in OpenStack: changes since Cactus and CloudPipe HA
Network in OpenStack: changes since Cactus and CloudPipe HA
 
Thinking outside the box, learning a little about a lot
Thinking outside the box, learning a little about a lotThinking outside the box, learning a little about a lot
Thinking outside the box, learning a little about a lot
 
Ac cuda c_3
Ac cuda c_3Ac cuda c_3
Ac cuda c_3
 
How to operate containerized OpenStack
How to operate containerized OpenStackHow to operate containerized OpenStack
How to operate containerized OpenStack
 

Similar to XPDS13: Increasing XenServer's VM density - Jonathan Davies, Citrix

Ceph Day New York 2014: Ceph, a physical perspective
Ceph Day New York 2014: Ceph, a physical perspective Ceph Day New York 2014: Ceph, a physical perspective
Ceph Day New York 2014: Ceph, a physical perspective Ceph Community
 
ARM server, The Cy7 Introduction by Aaron Joue, Ambedded Technology
ARM server, The Cy7 Introduction by Aaron Joue, Ambedded TechnologyARM server, The Cy7 Introduction by Aaron Joue, Ambedded Technology
ARM server, The Cy7 Introduction by Aaron Joue, Ambedded TechnologyAaron Joue
 
Preventing Possible PVS Performance Pain Points
Preventing Possible PVS Performance Pain PointsPreventing Possible PVS Performance Pain Points
Preventing Possible PVS Performance Pain PointsAndrew Wood
 
XPDS14 - Scaling Xen's Aggregate Storage Performance - Felipe Franciosi, Citrix
XPDS14 - Scaling Xen's Aggregate Storage Performance - Felipe Franciosi, CitrixXPDS14 - Scaling Xen's Aggregate Storage Performance - Felipe Franciosi, Citrix
XPDS14 - Scaling Xen's Aggregate Storage Performance - Felipe Franciosi, CitrixThe Linux Foundation
 
VMworld Europe 2014: Virtual SAN Best Practices and Use Cases
VMworld Europe 2014: Virtual SAN Best Practices and Use CasesVMworld Europe 2014: Virtual SAN Best Practices and Use Cases
VMworld Europe 2014: Virtual SAN Best Practices and Use CasesVMworld
 
Building a Stretched Cluster using Virtual SAN 6.1
Building a Stretched Cluster using Virtual SAN 6.1Building a Stretched Cluster using Virtual SAN 6.1
Building a Stretched Cluster using Virtual SAN 6.1Duncan Epping
 
VMworld 2013: vSphere Distributed Switch – Design and Best Practices
VMworld 2013: vSphere Distributed Switch – Design and Best Practices VMworld 2013: vSphere Distributed Switch – Design and Best Practices
VMworld 2013: vSphere Distributed Switch – Design and Best Practices VMworld
 
VMworld 2013: Successfully Virtualize Microsoft Exchange Server
VMworld 2013: Successfully Virtualize Microsoft Exchange Server VMworld 2013: Successfully Virtualize Microsoft Exchange Server
VMworld 2013: Successfully Virtualize Microsoft Exchange Server VMworld
 
Fusion-IO - Building a High Performance and Reliable VSAN Environment
Fusion-IO - Building a High Performance and Reliable VSAN EnvironmentFusion-IO - Building a High Performance and Reliable VSAN Environment
Fusion-IO - Building a High Performance and Reliable VSAN EnvironmentVMUG IT
 
SYN 104: Citrix and Nutanix
SYN 104: Citrix and Nutanix SYN 104: Citrix and Nutanix
SYN 104: Citrix and Nutanix Citrix
 
High Performance Linux Virtual Machine on Microsoft Azure: SR-IOV Networking ...
High Performance Linux Virtual Machine on Microsoft Azure: SR-IOV Networking ...High Performance Linux Virtual Machine on Microsoft Azure: SR-IOV Networking ...
High Performance Linux Virtual Machine on Microsoft Azure: SR-IOV Networking ...LinuxCon ContainerCon CloudOpen China
 
VMworld 2013: Enterprise Architecture Design for VMware Horizon View 5.2
VMworld 2013: Enterprise Architecture Design for VMware Horizon View 5.2 VMworld 2013: Enterprise Architecture Design for VMware Horizon View 5.2
VMworld 2013: Enterprise Architecture Design for VMware Horizon View 5.2 VMworld
 
FalconStor NSS Presentation
FalconStor NSS PresentationFalconStor NSS Presentation
FalconStor NSS Presentationrpsprowl
 
SRV402 Deep Dive on Amazon EC2 Instances, Featuring Performance Optimization ...
SRV402 Deep Dive on Amazon EC2 Instances, Featuring Performance Optimization ...SRV402 Deep Dive on Amazon EC2 Instances, Featuring Performance Optimization ...
SRV402 Deep Dive on Amazon EC2 Instances, Featuring Performance Optimization ...Amazon Web Services
 
Windows 8 news for the enterprise
Windows 8 news for the enterpriseWindows 8 news for the enterprise
Windows 8 news for the enterpriseOlav Tvedt
 
Journey to Stability: Petabyte Ceph Cluster in OpenStack Cloud
Journey to Stability: Petabyte Ceph Cluster in OpenStack CloudJourney to Stability: Petabyte Ceph Cluster in OpenStack Cloud
Journey to Stability: Petabyte Ceph Cluster in OpenStack CloudCeph Community
 
Journey to Stability: Petabyte Ceph Cluster in OpenStack Cloud
Journey to Stability: Petabyte Ceph Cluster in OpenStack CloudJourney to Stability: Petabyte Ceph Cluster in OpenStack Cloud
Journey to Stability: Petabyte Ceph Cluster in OpenStack CloudPatrick McGarry
 
DataStax: Extreme Cassandra Optimization: The Sequel
DataStax: Extreme Cassandra Optimization: The SequelDataStax: Extreme Cassandra Optimization: The Sequel
DataStax: Extreme Cassandra Optimization: The SequelDataStax Academy
 

Similar to XPDS13: Increasing XenServer's VM density - Jonathan Davies, Citrix (20)

Ceph Day New York 2014: Ceph, a physical perspective
Ceph Day New York 2014: Ceph, a physical perspective Ceph Day New York 2014: Ceph, a physical perspective
Ceph Day New York 2014: Ceph, a physical perspective
 
ARM server, The Cy7 Introduction by Aaron Joue, Ambedded Technology
ARM server, The Cy7 Introduction by Aaron Joue, Ambedded TechnologyARM server, The Cy7 Introduction by Aaron Joue, Ambedded Technology
ARM server, The Cy7 Introduction by Aaron Joue, Ambedded Technology
 
Preventing Possible PVS Performance Pain Points
Preventing Possible PVS Performance Pain PointsPreventing Possible PVS Performance Pain Points
Preventing Possible PVS Performance Pain Points
 
XPDS14 - Scaling Xen's Aggregate Storage Performance - Felipe Franciosi, Citrix
XPDS14 - Scaling Xen's Aggregate Storage Performance - Felipe Franciosi, CitrixXPDS14 - Scaling Xen's Aggregate Storage Performance - Felipe Franciosi, Citrix
XPDS14 - Scaling Xen's Aggregate Storage Performance - Felipe Franciosi, Citrix
 
Mellanox Storage Solutions
Mellanox Storage SolutionsMellanox Storage Solutions
Mellanox Storage Solutions
 
VMworld Europe 2014: Virtual SAN Best Practices and Use Cases
VMworld Europe 2014: Virtual SAN Best Practices and Use CasesVMworld Europe 2014: Virtual SAN Best Practices and Use Cases
VMworld Europe 2014: Virtual SAN Best Practices and Use Cases
 
Building a Stretched Cluster using Virtual SAN 6.1
Building a Stretched Cluster using Virtual SAN 6.1Building a Stretched Cluster using Virtual SAN 6.1
Building a Stretched Cluster using Virtual SAN 6.1
 
VMworld 2013: vSphere Distributed Switch – Design and Best Practices
VMworld 2013: vSphere Distributed Switch – Design and Best Practices VMworld 2013: vSphere Distributed Switch – Design and Best Practices
VMworld 2013: vSphere Distributed Switch – Design and Best Practices
 
VMworld 2013: Successfully Virtualize Microsoft Exchange Server
VMworld 2013: Successfully Virtualize Microsoft Exchange Server VMworld 2013: Successfully Virtualize Microsoft Exchange Server
VMworld 2013: Successfully Virtualize Microsoft Exchange Server
 
Fusion-IO - Building a High Performance and Reliable VSAN Environment
Fusion-IO - Building a High Performance and Reliable VSAN EnvironmentFusion-IO - Building a High Performance and Reliable VSAN Environment
Fusion-IO - Building a High Performance and Reliable VSAN Environment
 
SYN 104: Citrix and Nutanix
SYN 104: Citrix and Nutanix SYN 104: Citrix and Nutanix
SYN 104: Citrix and Nutanix
 
Implementing dr w. hyper v clustering
Implementing dr w. hyper v clusteringImplementing dr w. hyper v clustering
Implementing dr w. hyper v clustering
 
High Performance Linux Virtual Machine on Microsoft Azure: SR-IOV Networking ...
High Performance Linux Virtual Machine on Microsoft Azure: SR-IOV Networking ...High Performance Linux Virtual Machine on Microsoft Azure: SR-IOV Networking ...
High Performance Linux Virtual Machine on Microsoft Azure: SR-IOV Networking ...
 
VMworld 2013: Enterprise Architecture Design for VMware Horizon View 5.2
VMworld 2013: Enterprise Architecture Design for VMware Horizon View 5.2 VMworld 2013: Enterprise Architecture Design for VMware Horizon View 5.2
VMworld 2013: Enterprise Architecture Design for VMware Horizon View 5.2
 
FalconStor NSS Presentation
FalconStor NSS PresentationFalconStor NSS Presentation
FalconStor NSS Presentation
 
SRV402 Deep Dive on Amazon EC2 Instances, Featuring Performance Optimization ...
SRV402 Deep Dive on Amazon EC2 Instances, Featuring Performance Optimization ...SRV402 Deep Dive on Amazon EC2 Instances, Featuring Performance Optimization ...
SRV402 Deep Dive on Amazon EC2 Instances, Featuring Performance Optimization ...
 
Windows 8 news for the enterprise
Windows 8 news for the enterpriseWindows 8 news for the enterprise
Windows 8 news for the enterprise
 
Journey to Stability: Petabyte Ceph Cluster in OpenStack Cloud
Journey to Stability: Petabyte Ceph Cluster in OpenStack CloudJourney to Stability: Petabyte Ceph Cluster in OpenStack Cloud
Journey to Stability: Petabyte Ceph Cluster in OpenStack Cloud
 
Journey to Stability: Petabyte Ceph Cluster in OpenStack Cloud
Journey to Stability: Petabyte Ceph Cluster in OpenStack CloudJourney to Stability: Petabyte Ceph Cluster in OpenStack Cloud
Journey to Stability: Petabyte Ceph Cluster in OpenStack Cloud
 
DataStax: Extreme Cassandra Optimization: The Sequel
DataStax: Extreme Cassandra Optimization: The SequelDataStax: Extreme Cassandra Optimization: The Sequel
DataStax: Extreme Cassandra Optimization: The Sequel
 

More from The Linux Foundation

ELC2019: Static Partitioning Made Simple
ELC2019: Static Partitioning Made SimpleELC2019: Static Partitioning Made Simple
ELC2019: Static Partitioning Made SimpleThe Linux Foundation
 
XPDDS19: How TrenchBoot is Enabling Measured Launch for Open-Source Platform ...
XPDDS19: How TrenchBoot is Enabling Measured Launch for Open-Source Platform ...XPDDS19: How TrenchBoot is Enabling Measured Launch for Open-Source Platform ...
XPDDS19: How TrenchBoot is Enabling Measured Launch for Open-Source Platform ...The Linux Foundation
 
XPDDS19 Keynote: Xen in Automotive - Artem Mygaiev, Director, Technology Solu...
XPDDS19 Keynote: Xen in Automotive - Artem Mygaiev, Director, Technology Solu...XPDDS19 Keynote: Xen in Automotive - Artem Mygaiev, Director, Technology Solu...
XPDDS19 Keynote: Xen in Automotive - Artem Mygaiev, Director, Technology Solu...The Linux Foundation
 
XPDDS19 Keynote: Xen Project Weather Report 2019 - Lars Kurth, Director of Op...
XPDDS19 Keynote: Xen Project Weather Report 2019 - Lars Kurth, Director of Op...XPDDS19 Keynote: Xen Project Weather Report 2019 - Lars Kurth, Director of Op...
XPDDS19 Keynote: Xen Project Weather Report 2019 - Lars Kurth, Director of Op...The Linux Foundation
 
XPDDS19 Keynote: Unikraft Weather Report
XPDDS19 Keynote:  Unikraft Weather ReportXPDDS19 Keynote:  Unikraft Weather Report
XPDDS19 Keynote: Unikraft Weather ReportThe Linux Foundation
 
XPDDS19 Keynote: Secret-free Hypervisor: Now and Future - Wei Liu, Software E...
XPDDS19 Keynote: Secret-free Hypervisor: Now and Future - Wei Liu, Software E...XPDDS19 Keynote: Secret-free Hypervisor: Now and Future - Wei Liu, Software E...
XPDDS19 Keynote: Secret-free Hypervisor: Now and Future - Wei Liu, Software E...The Linux Foundation
 
XPDDS19 Keynote: Xen Dom0-less - Stefano Stabellini, Principal Engineer, Xilinx
XPDDS19 Keynote: Xen Dom0-less - Stefano Stabellini, Principal Engineer, XilinxXPDDS19 Keynote: Xen Dom0-less - Stefano Stabellini, Principal Engineer, Xilinx
XPDDS19 Keynote: Xen Dom0-less - Stefano Stabellini, Principal Engineer, XilinxThe Linux Foundation
 
XPDDS19 Keynote: Patch Review for Non-maintainers - George Dunlap, Citrix Sys...
XPDDS19 Keynote: Patch Review for Non-maintainers - George Dunlap, Citrix Sys...XPDDS19 Keynote: Patch Review for Non-maintainers - George Dunlap, Citrix Sys...
XPDDS19 Keynote: Patch Review for Non-maintainers - George Dunlap, Citrix Sys...The Linux Foundation
 
XPDDS19: Memories of a VM Funk - Mihai Donțu, Bitdefender
XPDDS19: Memories of a VM Funk - Mihai Donțu, BitdefenderXPDDS19: Memories of a VM Funk - Mihai Donțu, Bitdefender
XPDDS19: Memories of a VM Funk - Mihai Donțu, BitdefenderThe Linux Foundation
 
OSSJP/ALS19: The Road to Safety Certification: Overcoming Community Challeng...
OSSJP/ALS19:  The Road to Safety Certification: Overcoming Community Challeng...OSSJP/ALS19:  The Road to Safety Certification: Overcoming Community Challeng...
OSSJP/ALS19: The Road to Safety Certification: Overcoming Community Challeng...The Linux Foundation
 
OSSJP/ALS19: The Road to Safety Certification: How the Xen Project is Making...
 OSSJP/ALS19: The Road to Safety Certification: How the Xen Project is Making... OSSJP/ALS19: The Road to Safety Certification: How the Xen Project is Making...
OSSJP/ALS19: The Road to Safety Certification: How the Xen Project is Making...The Linux Foundation
 
XPDDS19: Speculative Sidechannels and Mitigations - Andrew Cooper, Citrix
XPDDS19: Speculative Sidechannels and Mitigations - Andrew Cooper, CitrixXPDDS19: Speculative Sidechannels and Mitigations - Andrew Cooper, Citrix
XPDDS19: Speculative Sidechannels and Mitigations - Andrew Cooper, CitrixThe Linux Foundation
 
XPDDS19: Keeping Coherency on Arm: Reborn - Julien Grall, Arm ltd
XPDDS19: Keeping Coherency on Arm: Reborn - Julien Grall, Arm ltdXPDDS19: Keeping Coherency on Arm: Reborn - Julien Grall, Arm ltd
XPDDS19: Keeping Coherency on Arm: Reborn - Julien Grall, Arm ltdThe Linux Foundation
 
XPDDS19: QEMU PV Backend 'qdevification'... What Does it Mean? - Paul Durrant...
XPDDS19: QEMU PV Backend 'qdevification'... What Does it Mean? - Paul Durrant...XPDDS19: QEMU PV Backend 'qdevification'... What Does it Mean? - Paul Durrant...
XPDDS19: QEMU PV Backend 'qdevification'... What Does it Mean? - Paul Durrant...The Linux Foundation
 
XPDDS19: Status of PCI Emulation in Xen - Roger Pau Monné, Citrix Systems R&D
XPDDS19: Status of PCI Emulation in Xen - Roger Pau Monné, Citrix Systems R&DXPDDS19: Status of PCI Emulation in Xen - Roger Pau Monné, Citrix Systems R&D
XPDDS19: Status of PCI Emulation in Xen - Roger Pau Monné, Citrix Systems R&DThe Linux Foundation
 
XPDDS19: [ARM] OP-TEE Mediator in Xen - Volodymyr Babchuk, EPAM Systems
XPDDS19: [ARM] OP-TEE Mediator in Xen - Volodymyr Babchuk, EPAM SystemsXPDDS19: [ARM] OP-TEE Mediator in Xen - Volodymyr Babchuk, EPAM Systems
XPDDS19: [ARM] OP-TEE Mediator in Xen - Volodymyr Babchuk, EPAM SystemsThe Linux Foundation
 
XPDDS19: Bringing Xen to the Masses: The Story of Building a Community-driven...
XPDDS19: Bringing Xen to the Masses: The Story of Building a Community-driven...XPDDS19: Bringing Xen to the Masses: The Story of Building a Community-driven...
XPDDS19: Bringing Xen to the Masses: The Story of Building a Community-driven...The Linux Foundation
 
XPDDS19: Will Robots Automate Your Job Away? Streamlining Xen Project Contrib...
XPDDS19: Will Robots Automate Your Job Away? Streamlining Xen Project Contrib...XPDDS19: Will Robots Automate Your Job Away? Streamlining Xen Project Contrib...
XPDDS19: Will Robots Automate Your Job Away? Streamlining Xen Project Contrib...The Linux Foundation
 
XPDDS19: Client Virtualization Toolstack in Go - Nick Rosbrook & Brendan Kerr...
XPDDS19: Client Virtualization Toolstack in Go - Nick Rosbrook & Brendan Kerr...XPDDS19: Client Virtualization Toolstack in Go - Nick Rosbrook & Brendan Kerr...
XPDDS19: Client Virtualization Toolstack in Go - Nick Rosbrook & Brendan Kerr...The Linux Foundation
 
XPDDS19: Core Scheduling in Xen - Jürgen Groß, SUSE
XPDDS19: Core Scheduling in Xen - Jürgen Groß, SUSEXPDDS19: Core Scheduling in Xen - Jürgen Groß, SUSE
XPDDS19: Core Scheduling in Xen - Jürgen Groß, SUSEThe Linux Foundation
 

More from The Linux Foundation (20)

ELC2019: Static Partitioning Made Simple
ELC2019: Static Partitioning Made SimpleELC2019: Static Partitioning Made Simple
ELC2019: Static Partitioning Made Simple
 
XPDDS19: How TrenchBoot is Enabling Measured Launch for Open-Source Platform ...
XPDDS19: How TrenchBoot is Enabling Measured Launch for Open-Source Platform ...XPDDS19: How TrenchBoot is Enabling Measured Launch for Open-Source Platform ...
XPDDS19: How TrenchBoot is Enabling Measured Launch for Open-Source Platform ...
 
XPDDS19 Keynote: Xen in Automotive - Artem Mygaiev, Director, Technology Solu...
XPDDS19 Keynote: Xen in Automotive - Artem Mygaiev, Director, Technology Solu...XPDDS19 Keynote: Xen in Automotive - Artem Mygaiev, Director, Technology Solu...
XPDDS19 Keynote: Xen in Automotive - Artem Mygaiev, Director, Technology Solu...
 
XPDDS19 Keynote: Xen Project Weather Report 2019 - Lars Kurth, Director of Op...
XPDDS19 Keynote: Xen Project Weather Report 2019 - Lars Kurth, Director of Op...XPDDS19 Keynote: Xen Project Weather Report 2019 - Lars Kurth, Director of Op...
XPDDS19 Keynote: Xen Project Weather Report 2019 - Lars Kurth, Director of Op...
 
XPDDS19 Keynote: Unikraft Weather Report
XPDDS19 Keynote:  Unikraft Weather ReportXPDDS19 Keynote:  Unikraft Weather Report
XPDDS19 Keynote: Unikraft Weather Report
 
XPDDS19 Keynote: Secret-free Hypervisor: Now and Future - Wei Liu, Software E...
XPDDS19 Keynote: Secret-free Hypervisor: Now and Future - Wei Liu, Software E...XPDDS19 Keynote: Secret-free Hypervisor: Now and Future - Wei Liu, Software E...
XPDDS19 Keynote: Secret-free Hypervisor: Now and Future - Wei Liu, Software E...
 
XPDDS19 Keynote: Xen Dom0-less - Stefano Stabellini, Principal Engineer, Xilinx
XPDDS19 Keynote: Xen Dom0-less - Stefano Stabellini, Principal Engineer, XilinxXPDDS19 Keynote: Xen Dom0-less - Stefano Stabellini, Principal Engineer, Xilinx
XPDDS19 Keynote: Xen Dom0-less - Stefano Stabellini, Principal Engineer, Xilinx
 
XPDDS19 Keynote: Patch Review for Non-maintainers - George Dunlap, Citrix Sys...
XPDDS19 Keynote: Patch Review for Non-maintainers - George Dunlap, Citrix Sys...XPDDS19 Keynote: Patch Review for Non-maintainers - George Dunlap, Citrix Sys...
XPDDS19 Keynote: Patch Review for Non-maintainers - George Dunlap, Citrix Sys...
 
XPDDS19: Memories of a VM Funk - Mihai Donțu, Bitdefender
XPDDS19: Memories of a VM Funk - Mihai Donțu, BitdefenderXPDDS19: Memories of a VM Funk - Mihai Donțu, Bitdefender
XPDDS19: Memories of a VM Funk - Mihai Donțu, Bitdefender
 
OSSJP/ALS19: The Road to Safety Certification: Overcoming Community Challeng...
OSSJP/ALS19:  The Road to Safety Certification: Overcoming Community Challeng...OSSJP/ALS19:  The Road to Safety Certification: Overcoming Community Challeng...
OSSJP/ALS19: The Road to Safety Certification: Overcoming Community Challeng...
 
OSSJP/ALS19: The Road to Safety Certification: How the Xen Project is Making...
 OSSJP/ALS19: The Road to Safety Certification: How the Xen Project is Making... OSSJP/ALS19: The Road to Safety Certification: How the Xen Project is Making...
OSSJP/ALS19: The Road to Safety Certification: How the Xen Project is Making...
 
XPDDS19: Speculative Sidechannels and Mitigations - Andrew Cooper, Citrix
XPDDS19: Speculative Sidechannels and Mitigations - Andrew Cooper, CitrixXPDDS19: Speculative Sidechannels and Mitigations - Andrew Cooper, Citrix
XPDDS19: Speculative Sidechannels and Mitigations - Andrew Cooper, Citrix
 
XPDDS19: Keeping Coherency on Arm: Reborn - Julien Grall, Arm ltd
XPDDS19: Keeping Coherency on Arm: Reborn - Julien Grall, Arm ltdXPDDS19: Keeping Coherency on Arm: Reborn - Julien Grall, Arm ltd
XPDDS19: Keeping Coherency on Arm: Reborn - Julien Grall, Arm ltd
 
XPDDS19: QEMU PV Backend 'qdevification'... What Does it Mean? - Paul Durrant...
XPDDS19: QEMU PV Backend 'qdevification'... What Does it Mean? - Paul Durrant...XPDDS19: QEMU PV Backend 'qdevification'... What Does it Mean? - Paul Durrant...
XPDDS19: QEMU PV Backend 'qdevification'... What Does it Mean? - Paul Durrant...
 
XPDDS19: Status of PCI Emulation in Xen - Roger Pau Monné, Citrix Systems R&D
XPDDS19: Status of PCI Emulation in Xen - Roger Pau Monné, Citrix Systems R&DXPDDS19: Status of PCI Emulation in Xen - Roger Pau Monné, Citrix Systems R&D
XPDDS19: Status of PCI Emulation in Xen - Roger Pau Monné, Citrix Systems R&D
 
XPDDS19: [ARM] OP-TEE Mediator in Xen - Volodymyr Babchuk, EPAM Systems
XPDDS19: [ARM] OP-TEE Mediator in Xen - Volodymyr Babchuk, EPAM SystemsXPDDS19: [ARM] OP-TEE Mediator in Xen - Volodymyr Babchuk, EPAM Systems
XPDDS19: [ARM] OP-TEE Mediator in Xen - Volodymyr Babchuk, EPAM Systems
 
XPDDS19: Bringing Xen to the Masses: The Story of Building a Community-driven...
XPDDS19: Bringing Xen to the Masses: The Story of Building a Community-driven...XPDDS19: Bringing Xen to the Masses: The Story of Building a Community-driven...
XPDDS19: Bringing Xen to the Masses: The Story of Building a Community-driven...
 
XPDDS19: Will Robots Automate Your Job Away? Streamlining Xen Project Contrib...
XPDDS19: Will Robots Automate Your Job Away? Streamlining Xen Project Contrib...XPDDS19: Will Robots Automate Your Job Away? Streamlining Xen Project Contrib...
XPDDS19: Will Robots Automate Your Job Away? Streamlining Xen Project Contrib...
 
XPDDS19: Client Virtualization Toolstack in Go - Nick Rosbrook & Brendan Kerr...
XPDDS19: Client Virtualization Toolstack in Go - Nick Rosbrook & Brendan Kerr...XPDDS19: Client Virtualization Toolstack in Go - Nick Rosbrook & Brendan Kerr...
XPDDS19: Client Virtualization Toolstack in Go - Nick Rosbrook & Brendan Kerr...
 
XPDDS19: Core Scheduling in Xen - Jürgen Groß, SUSE
XPDDS19: Core Scheduling in Xen - Jürgen Groß, SUSEXPDDS19: Core Scheduling in Xen - Jürgen Groß, SUSE
XPDDS19: Core Scheduling in Xen - Jürgen Groß, SUSE
 

Recently uploaded

Dev Dives: Streamline document processing with UiPath Studio Web
Dev Dives: Streamline document processing with UiPath Studio WebDev Dives: Streamline document processing with UiPath Studio Web
Dev Dives: Streamline document processing with UiPath Studio WebUiPathCommunity
 
Search Engine Optimization SEO PDF for 2024.pdf
Search Engine Optimization SEO PDF for 2024.pdfSearch Engine Optimization SEO PDF for 2024.pdf
Search Engine Optimization SEO PDF for 2024.pdfRankYa
 
Are Multi-Cloud and Serverless Good or Bad?
Are Multi-Cloud and Serverless Good or Bad?Are Multi-Cloud and Serverless Good or Bad?
Are Multi-Cloud and Serverless Good or Bad?Mattias Andersson
 
H2O.ai CEO/Founder: Sri Ambati Keynote at Wells Fargo Day
H2O.ai CEO/Founder: Sri Ambati Keynote at Wells Fargo DayH2O.ai CEO/Founder: Sri Ambati Keynote at Wells Fargo Day
H2O.ai CEO/Founder: Sri Ambati Keynote at Wells Fargo DaySri Ambati
 
SAP Build Work Zone - Overview L2-L3.pptx
SAP Build Work Zone - Overview L2-L3.pptxSAP Build Work Zone - Overview L2-L3.pptx
SAP Build Work Zone - Overview L2-L3.pptxNavinnSomaal
 
Hyperautomation and AI/ML: A Strategy for Digital Transformation Success.pdf
Hyperautomation and AI/ML: A Strategy for Digital Transformation Success.pdfHyperautomation and AI/ML: A Strategy for Digital Transformation Success.pdf
Hyperautomation and AI/ML: A Strategy for Digital Transformation Success.pdfPrecisely
 
TrustArc Webinar - How to Build Consumer Trust Through Data Privacy
TrustArc Webinar - How to Build Consumer Trust Through Data PrivacyTrustArc Webinar - How to Build Consumer Trust Through Data Privacy
TrustArc Webinar - How to Build Consumer Trust Through Data PrivacyTrustArc
 
"ML in Production",Oleksandr Bagan
"ML in Production",Oleksandr Bagan"ML in Production",Oleksandr Bagan
"ML in Production",Oleksandr BaganFwdays
 
Leverage Zilliz Serverless - Up to 50X Saving for Your Vector Storage Cost
Leverage Zilliz Serverless - Up to 50X Saving for Your Vector Storage CostLeverage Zilliz Serverless - Up to 50X Saving for Your Vector Storage Cost
Leverage Zilliz Serverless - Up to 50X Saving for Your Vector Storage CostZilliz
 
Advanced Test Driven-Development @ php[tek] 2024
Advanced Test Driven-Development @ php[tek] 2024Advanced Test Driven-Development @ php[tek] 2024
Advanced Test Driven-Development @ php[tek] 2024Scott Keck-Warren
 
DevEX - reference for building teams, processes, and platforms
DevEX - reference for building teams, processes, and platformsDevEX - reference for building teams, processes, and platforms
DevEX - reference for building teams, processes, and platformsSergiu Bodiu
 
Streamlining Python Development: A Guide to a Modern Project Setup
Streamlining Python Development: A Guide to a Modern Project SetupStreamlining Python Development: A Guide to a Modern Project Setup
Streamlining Python Development: A Guide to a Modern Project SetupFlorian Wilhelm
 
Unraveling Multimodality with Large Language Models.pdf
Unraveling Multimodality with Large Language Models.pdfUnraveling Multimodality with Large Language Models.pdf
Unraveling Multimodality with Large Language Models.pdfAlex Barbosa Coqueiro
 
Human Factors of XR: Using Human Factors to Design XR Systems
Human Factors of XR: Using Human Factors to Design XR SystemsHuman Factors of XR: Using Human Factors to Design XR Systems
Human Factors of XR: Using Human Factors to Design XR SystemsMark Billinghurst
 
New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024
New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024
New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024BookNet Canada
 
Story boards and shot lists for my a level piece
Story boards and shot lists for my a level pieceStory boards and shot lists for my a level piece
Story boards and shot lists for my a level piececharlottematthew16
 
Merck Moving Beyond Passwords: FIDO Paris Seminar.pptx
Merck Moving Beyond Passwords: FIDO Paris Seminar.pptxMerck Moving Beyond Passwords: FIDO Paris Seminar.pptx
Merck Moving Beyond Passwords: FIDO Paris Seminar.pptxLoriGlavin3
 
Developer Data Modeling Mistakes: From Postgres to NoSQL
Developer Data Modeling Mistakes: From Postgres to NoSQLDeveloper Data Modeling Mistakes: From Postgres to NoSQL
Developer Data Modeling Mistakes: From Postgres to NoSQLScyllaDB
 
Vertex AI Gemini Prompt Engineering Tips
Vertex AI Gemini Prompt Engineering TipsVertex AI Gemini Prompt Engineering Tips
Vertex AI Gemini Prompt Engineering TipsMiki Katsuragi
 

Recently uploaded (20)

Dev Dives: Streamline document processing with UiPath Studio Web
Dev Dives: Streamline document processing with UiPath Studio WebDev Dives: Streamline document processing with UiPath Studio Web
Dev Dives: Streamline document processing with UiPath Studio Web
 
DMCC Future of Trade Web3 - Special Edition
DMCC Future of Trade Web3 - Special EditionDMCC Future of Trade Web3 - Special Edition
DMCC Future of Trade Web3 - Special Edition
 
Search Engine Optimization SEO PDF for 2024.pdf
Search Engine Optimization SEO PDF for 2024.pdfSearch Engine Optimization SEO PDF for 2024.pdf
Search Engine Optimization SEO PDF for 2024.pdf
 
Are Multi-Cloud and Serverless Good or Bad?
Are Multi-Cloud and Serverless Good or Bad?Are Multi-Cloud and Serverless Good or Bad?
Are Multi-Cloud and Serverless Good or Bad?
 
H2O.ai CEO/Founder: Sri Ambati Keynote at Wells Fargo Day
H2O.ai CEO/Founder: Sri Ambati Keynote at Wells Fargo DayH2O.ai CEO/Founder: Sri Ambati Keynote at Wells Fargo Day
H2O.ai CEO/Founder: Sri Ambati Keynote at Wells Fargo Day
 
SAP Build Work Zone - Overview L2-L3.pptx
SAP Build Work Zone - Overview L2-L3.pptxSAP Build Work Zone - Overview L2-L3.pptx
SAP Build Work Zone - Overview L2-L3.pptx
 
Hyperautomation and AI/ML: A Strategy for Digital Transformation Success.pdf
Hyperautomation and AI/ML: A Strategy for Digital Transformation Success.pdfHyperautomation and AI/ML: A Strategy for Digital Transformation Success.pdf
Hyperautomation and AI/ML: A Strategy for Digital Transformation Success.pdf
 
TrustArc Webinar - How to Build Consumer Trust Through Data Privacy
TrustArc Webinar - How to Build Consumer Trust Through Data PrivacyTrustArc Webinar - How to Build Consumer Trust Through Data Privacy
TrustArc Webinar - How to Build Consumer Trust Through Data Privacy
 
"ML in Production",Oleksandr Bagan
"ML in Production",Oleksandr Bagan"ML in Production",Oleksandr Bagan
"ML in Production",Oleksandr Bagan
 
Leverage Zilliz Serverless - Up to 50X Saving for Your Vector Storage Cost
Leverage Zilliz Serverless - Up to 50X Saving for Your Vector Storage CostLeverage Zilliz Serverless - Up to 50X Saving for Your Vector Storage Cost
Leverage Zilliz Serverless - Up to 50X Saving for Your Vector Storage Cost
 
Advanced Test Driven-Development @ php[tek] 2024
Advanced Test Driven-Development @ php[tek] 2024Advanced Test Driven-Development @ php[tek] 2024
Advanced Test Driven-Development @ php[tek] 2024
 
DevEX - reference for building teams, processes, and platforms
DevEX - reference for building teams, processes, and platformsDevEX - reference for building teams, processes, and platforms
DevEX - reference for building teams, processes, and platforms
 
Streamlining Python Development: A Guide to a Modern Project Setup
Streamlining Python Development: A Guide to a Modern Project SetupStreamlining Python Development: A Guide to a Modern Project Setup
Streamlining Python Development: A Guide to a Modern Project Setup
 
Unraveling Multimodality with Large Language Models.pdf
Unraveling Multimodality with Large Language Models.pdfUnraveling Multimodality with Large Language Models.pdf
Unraveling Multimodality with Large Language Models.pdf
 
Human Factors of XR: Using Human Factors to Design XR Systems
Human Factors of XR: Using Human Factors to Design XR SystemsHuman Factors of XR: Using Human Factors to Design XR Systems
Human Factors of XR: Using Human Factors to Design XR Systems
 
New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024
New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024
New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024
 
Story boards and shot lists for my a level piece
Story boards and shot lists for my a level pieceStory boards and shot lists for my a level piece
Story boards and shot lists for my a level piece
 
Merck Moving Beyond Passwords: FIDO Paris Seminar.pptx
Merck Moving Beyond Passwords: FIDO Paris Seminar.pptxMerck Moving Beyond Passwords: FIDO Paris Seminar.pptx
Merck Moving Beyond Passwords: FIDO Paris Seminar.pptx
 
Developer Data Modeling Mistakes: From Postgres to NoSQL
Developer Data Modeling Mistakes: From Postgres to NoSQLDeveloper Data Modeling Mistakes: From Postgres to NoSQL
Developer Data Modeling Mistakes: From Postgres to NoSQL
 
Vertex AI Gemini Prompt Engineering Tips
Vertex AI Gemini Prompt Engineering TipsVertex AI Gemini Prompt Engineering Tips
Vertex AI Gemini Prompt Engineering Tips
 

XPDS13: Increasing XenServer's VM density - Jonathan Davies, Citrix

  • 1. Increasing XenServer’s VM density Jonathan Davies, XenServer System Performance Lead XenServer Engineering, Citrix Cambridge, UK 24 Oct 2013 Jonathan Davies (Citrix) Increasing XenServer’s VM density 24 Oct 2013 1 / 34
  • 2. Outline 1 Scalability expectations 2 Hard limits 3 Soft limits 4 Benchmarks Jonathan Davies (Citrix) Increasing XenServer’s VM density 24 Oct 2013 2 / 34
  • 3. Scalability expectations Outline 1 Scalability expectations 2 Hard limits 3 Soft limits 4 Benchmarks Jonathan Davies (Citrix) Increasing XenServer’s VM density 24 Oct 2013 3 / 34
  • 4. Scalability expectations Users expect VM density to scale with hardware Jonathan Davies (Citrix) Increasing XenServer’s VM density 24 Oct 2013 4 / 34
  • 5. Scalability expectations Users expect VM density to scale with hardware Jonathan Davies (Citrix) Increasing XenServer’s VM density 24 Oct 2013 5 / 34
  • 6. Scalability expectations Users get upset when. . . Jonathan Davies (Citrix) Increasing XenServer’s VM density 24 Oct 2013 6 / 34
  • 7. Scalability expectations XenServer’s VM density scalability hard density limit :-( hardware's theoretical capacity XS 6.1 (and earlier) Jonathan Davies (Citrix) Increasing XenServer’s VM density 24 Oct 2013 7 / 34
  • 8. Scalability expectations XenServer’s VM density scalability hard density limit :-( hardware's theoretical capacity XS 6.1 (and earlier) XS 6.2 practical density limit (depending on nature of VMs) Jonathan Davies (Citrix) Increasing XenServer’s VM density hard density limit :-) 24 Oct 2013 8 / 34
  • 9. Hard limits Outline 1 Scalability expectations 2 Hard limits 3 Soft limits 4 Benchmarks Jonathan Davies (Citrix) Increasing XenServer’s VM density 24 Oct 2013 9 / 34
  • 10. Hard limits Enumerated causes of limitations Hard limit 1: dom0 event channels Cause of limitation XenServer uses a 32-bit dom0 This means 1,024 dom0 event channels #define MAX_EVTCHNS(d) (BITS_PER_EVTCHN_WORD(d) * BITS_PER_EVTCHN_WORD(d)) Various VM functions use a dom0 event channel VM density hard limit 225 VMs per host (PV with 1 vCPU, 1 VIF, 1 VBD) 150 VMs per host (HVM with 1 vCPU, 1 VIF, 3 VBDs) Mitigation for XS 6.2 Hack for dom0 to enjoy 4,096 event channels → 800 VMs per host (PV with 1 vCPU, 1 VIF, 1 VBD) → 570 VMs per host (HVM with 1 vCPU, 1 VIF, 3 VBDs) Mitigation for future Change the ABI to provide unlimited event channels → this would remove the limit Jonathan Davies (Citrix) Increasing XenServer’s VM density 24 Oct 2013 10 / 34
  • 11. Hard limits Enumerated causes of limitations Hard limit 2: blktap2 device minor numbers Cause of limitation blktap2 only supports up to 1,024 minor numbers (despite the kernel allowing up to 1,048,576) #define MAX_BLKTAP_DEVICE 1024 Each virtual block device requires one device VM density hard limit 341 VMs per host (with 3 disks per VM) Mitigation for XS 6.2 Double this constant to 2,048 → 682 VMs per host (with 3 disks per VM) Mitigation for future Move away from blktap2 altogether? → this would remove the limit Jonathan Davies (Citrix) Increasing XenServer’s VM density 24 Oct 2013 11 / 34
  • 12. Hard limits Enumerated causes of limitations Hard limit 3: number of aio requests Cause of limitation Each blktap2 instance creates an asynchronous I/O context for receiving 402 events. Default system-wide number of aio requests was 444,416 in XS 6.1. VM density hard limit 368 VMs per host (with 3 disks per VM) Mitigation for XS 6.2 Set fs.aio-max-nr to 1,048,576 → 869 VMs per host (with 3 disks per VM) Mitigation for future Increase fs.aio-max-nr further or use storage driver domains → this would remove the limit Jonathan Davies (Citrix) Increasing XenServer’s VM density 24 Oct 2013 12 / 34
  • 13. Hard limits Enumerated causes of limitations Hard limit 4: dom0 grant references Cause of limitation Windows VMs use receive-side copy (RSC) by default in XS 6.1. netback allocates (at least) 22 grant-table entries per virtual interface for RSC. dom0 had a total of 8,192 grant-table entries in XS 6.1. VM density hard limit 372 VMs per host (with 1 interface per VM) Mitigation for XS 6.2 Don’t use RSC in Windows VMs anyway → this removes the limit Jonathan Davies (Citrix) Increasing XenServer’s VM density 24 Oct 2013 13 / 34
  • 14. Hard limits Enumerated causes of limitations Hard limit 5: connections to xenstored Cause of limitation xenstored uses select(2), which can only listen on 1,024 file descriptors. #define __FD_SETSIZE 1024 qemu opens 3 file descriptors to xenstored. VM density hard limit 333 VMs per host (HVM) Mitigation for XS 6.2 Make two qemu watches share a connection → 500 VMs per host (HVM) Mitigation for future Upstream qemu doesn’t connect to xenstored → this will remove the limit Jonathan Davies (Citrix) Increasing XenServer’s VM density 24 Oct 2013 14 / 34
  • 15. Hard limits Enumerated causes of limitations Hard limit 6: connections to consoled Cause of limitation Similarly, consoled uses select(2) Each PV domain opens 3 fds to consoled VM density hard limit 341 VMs per host (PV) Mitigation for XS 6.2 Use poll(2) rather than select(2) in consoled → this removes the limit Jonathan Davies (Citrix) Increasing XenServer’s VM density 24 Oct 2013 15 / 34
  • 16. Hard limits Enumerated causes of limitations Hard limit 7: dom0 low memory Cause of limitation Each running VM eats about 1MB of dom0 lowmem VM density hard limit around 650 VMs per host Mitigation for future Use a 64-bit dom0 → this will remove the limit Jonathan Davies (Citrix) Increasing XenServer’s VM density 24 Oct 2013 16 / 34
  • 17. Hard limits Summary of hard limits Summary of hard limits Limits on number of HVM guests with 1 vCPU, 1 VBD, 1 VIF (with PV drivers) Limitation dom0 event channels blktap minor numbers aio requests dom0 grant references xenstored connections consoled connections dom0 low memory Overall limit Limited by Jonathan Davies (Citrix) XS 6.1 225 1024 1105 372 333 no limit 650 225 event channels XS 6.2 800 2048 2608 no limit 500 no limit 650 500 xenstored Increasing XenServer’s VM density Future no limit no limit no limit no limit no limit no limit no limit very high something else! 24 Oct 2013 17 / 34
  • 18. Hard limits Summary of hard limits Summary of hard limits Limits on number of HVM guests with 1 vCPU, 3 VBDs, 1 VIF (with PV drivers) Limitation dom0 event channels blktap minor numbers aio requests dom0 grant references xenstored connections consoled connections dom0 low memory Overall limit Limited by Jonathan Davies (Citrix) XS 6.1 150 341 368 372 333 no limit 650 150 event channels XS 6.2 570 682 869 no limit 500 no limit 650 500 xenstored Increasing XenServer’s VM density Future no limit no limit no limit no limit no limit no limit no limit very high something else! 24 Oct 2013 18 / 34
  • 19. Hard limits Summary of hard limits Summary of hard limits Limits on number of PV guests with 1 vCPU, 1 VBD, 1 VIF Limitation dom0 event channels blktap minor numbers aio requests dom0 grant references xenstored connections consoled connections dom0 low memory Overall limit Limited by Jonathan Davies (Citrix) XS 6.1 225 1024 368 no limit no limit 341 650 225 event channels XS 6.2 1000 2048 869 no limit no limit no limit 650 650 dom0 lowmem Increasing XenServer’s VM density Future no limit no limit no limit no limit no limit no limit no limit very high something else! 24 Oct 2013 19 / 34
  • 20. Hard limits Summary of hard limits 500 Windows VMs on a host Jonathan Davies (Citrix) Increasing XenServer’s VM density 24 Oct 2013 20 / 34
  • 21. Soft limits Outline 1 Scalability expectations 2 Hard limits 3 Soft limits 4 Benchmarks Jonathan Davies (Citrix) Increasing XenServer’s VM density 24 Oct 2013 21 / 34
  • 22. Soft limits xenstored High dom0 CPU utilisation by xenstored top - 16:29:33 up 36 min, 1 user, load average: 0.80, 0.56, 0.47 Tasks: 132 total, 1 running, 131 sleeping, 0 stopped, 0 zombie Cpu(s): 40.1%us, 40.0%sy, 0.0%ni, 17.6%id, 0.0%wa, 0.0%hi, 0.0%si, 0.0%st Mem: 4186504k total, 443480k used, 3743024k free, 23696k buffers Swap: 524280k total, 0k used, 524280k free, 132504k cached PID 7339 6665 7225 7269 7413 7283 10938 16403 1 2 3 USER root root root root root root root root root root root PR 20 20 20 20 20 20 20 20 20 20 RT Jonathan Davies (Citrix) NI VIRT RES SHR S %CPU %MEM 0 6732 2240 840 S 80.2 0.1 0 4344 2636 584 S 0.4 0.1 0 48892 5356 1736 S 0.3 0.1 0 23704 3684 1308 S 0.3 0.1 0 195m 21m 8932 S 0.3 0.5 0 7492 4860 1200 S 0.3 0.1 0 29808 1856 956 S 0.3 0.0 0 2428 1104 824 R 0.3 0.0 0 2164 656 564 S 0.0 0.0 0 0 0 0 S 0.0 0.0 0 0 0 0 S 0.0 0.0 Increasing XenServer’s VM density TIME+ 0:10.22 0:04.03 0:03.35 0:03.47 0:10.28 0:08.65 0:00.40 0:02.31 0:00.83 0:00.00 0:00.01 COMMAND xenstored fe xcp-rrdd xcp-rrdd-iostat xapi xcp-rrdd-xenpm v6d top init kthreadd migration/0 24 Oct 2013 22 / 34
  • 23. Soft limits xenstored High dom0 CPU utilisation by xenstored dom0 vCPUs domU vCPUs Jonathan Davies (Citrix) Increasing XenServer’s VM density 24 Oct 2013 23 / 34
  • 24. Soft limits xenstored High dom0 CPU utilisation by xenstored xenstored's dom0 vCPU other dom0 vCPUs domU vCPUs Jonathan Davies (Citrix) Increasing XenServer’s VM density 24 Oct 2013 24 / 34
  • 25. Soft limits xenstored High dom0 CPU utilisation by xenstored Cause of limitation xenstored CPU utilisation bottleneck Mitigation for XS 6.2 Reduce xenstore use by XenServer’s toolstack: remove some spurious writes replace polling with watching Jonathan Davies (Citrix) Increasing XenServer’s VM density 24 Oct 2013 25 / 34
  • 26. Soft limits qemu High dom0 CPU utilisation due to qemu top - 16:40:27 up 2:07, 1 user, load average: 89.62, 87.22, 76.90 Tasks: 1015 total, 65 running, 950 sleeping, 0 stopped, 0 zombie Cpu(s): 23.4%us, 55.5%sy, 0.0%ni, 4.8%id, 0.0%wa, 0.0%hi, 15.4%si, 0.5%st Mem: 4180480k total, 1615840k used, 2564640k free, 3804k buffers Swap: 524280k total, 0k used, 524280k free, 122852k cached PID 7143 6653 16496 16970 997 3195 3497 3562 3993 7597 8150 8679 8974 11937 12545 14053 17752 USER root root tcpdump root 65583 65684 65656 65685 65546 65659 65550 65627 65661 root 65556 65598 65567 PR NI VIRT RES SHR S 20 0 0 0 0 R 10 -10 12264 7796 1152 R 20 0 5508 2132 1248 R 20 0 2952 1552 736 R 20 0 24696 4732 1572 S 20 0 24632 4736 1572 S 20 0 24760 4740 1576 R 20 0 24696 4732 1572 S 20 0 24888 4744 1580 S 20 0 24632 4736 1576 S 20 0 24760 4740 1580 R 20 0 24632 4740 1576 R 20 0 24568 4736 1572 S 20 0 0 0 0 S 20 0 24824 4748 1584 S 20 0 24760 4736 1576 S 20 0 24952 4740 1576 S Jonathan Davies (Citrix) %CPU %MEM 33.9 0.0 31.8 0.2 10.5 0.1 6.3 0.0 3.1 0.1 3.1 0.1 3.1 0.1 3.1 0.1 3.1 0.1 3.1 0.1 3.1 0.1 3.1 0.1 3.1 0.1 3.1 0.0 3.1 0.1 3.1 0.1 3.1 0.1 Increasing XenServer’s VM density TIME+ 17:21.63 36:14.34 5:35.12 0:00.11 0:56.30 0:27.34 0:28.65 0:26.97 0:53.19 0:28.86 0:51.71 0:31.18 0:27.97 1:12.92 0:58.46 0:31.33 0:56.82 COMMAND rpciod/0 ovs-vswitchd tcpdump top qemu-dm qemu-dm qemu-dm qemu-dm qemu-dm qemu-dm qemu-dm qemu-dm qemu-dm nfsiod qemu-dm qemu-dm qemu-dm 24 Oct 2013 26 / 34
  • 27. Soft limits qemu qemu burning dom0 CPU 200 idle Windows guests, each qemu utilising 3% of a CPU means 6 dom0 vCPUs wasted! Jonathan Davies (Citrix) Increasing XenServer’s VM density 24 Oct 2013 27 / 34
  • 28. Soft limits qemu What is qemu busy doing? Emulated device USB CD-ROM Buffered I/O & RTC timer Parallel port Serial port VNC qemu monitor qemu events per VM per second 221 38 13 1 1 1 1 Mitigation for XS 6.2 Use an event-channel for buffered I/O notifications Provide options to disable all emulated devices Jonathan Davies (Citrix) Increasing XenServer’s VM density 24 Oct 2013 28 / 34
  • 29. Benchmarks Outline 1 Scalability expectations 2 Hard limits 3 Soft limits 4 Benchmarks Jonathan Davies (Citrix) Increasing XenServer’s VM density 24 Oct 2013 29 / 34
  • 30. Benchmarks Bootstorm Booting 90 Win7 VMs Time to fully boot 90 VMs (25 at a time) 2500 Tampa (XS 6.1) Clearwater (XS 6.2) Elapsed time (s) 2000 1500 1000 XS 6.2 is 60% faster 500 0 0 25 50 75 100 125 150 175 200 VM index Machine used: Dell PowerEdge R815 (Quad 16-core AMD Opteron 6272 @ 2.1GHz) Jonathan Davies (Citrix) Increasing XenServer’s VM density 24 Oct 2013 30 / 34
  • 31. Benchmarks Bootstorm Booting 120 Win7 VMs Time to fully boot 120 VMs (25 at a time) 2500 Tampa (XS 6.1) Clearwater (XS 6.2) Elapsed time (s) 2000 1500 1000 XS 6.2 is 75% faster 500 0 0 25 50 75 100 125 150 175 200 VM index Machine used: Dell PowerEdge R815 (Quad 16-core AMD Opteron 6272 @ 2.1GHz) Jonathan Davies (Citrix) Increasing XenServer’s VM density 24 Oct 2013 31 / 34
  • 32. Benchmarks Bootstorm Booting 200 Win7 VMs Time to fully boot 200 VMs (25 at a time) 2500 Tampa (XS 6.1) Clearwater (XS 6.2) XS 6.1 can't even get 200 VMs running! Elapsed time (s) 2000 1500 It took XS 6.2 just 13 minutes to boot 200 VMs (on this hardware) 1000 500 0 0 25 50 75 100 125 150 175 200 VM index Machine used: Dell PowerEdge R815 (Quad 16-core AMD Opteron 6272 @ 2.1GHz) Jonathan Davies (Citrix) Increasing XenServer’s VM density 24 Oct 2013 32 / 34
  • 33. Benchmarks LoginVSI number of VMs performing acceptably LoginVSI: number of usable Windows VMs XS 6.2 XS 6.1 number of VMs running Machine used: Quad 10-core Intel E7-4860 @ 2.27GHz Jonathan Davies (Citrix) Increasing XenServer’s VM density 24 Oct 2013 33 / 34
  • 34. Questions Questions ? Jonathan Davies (Citrix) Increasing XenServer’s VM density 24 Oct 2013 34 / 34