Mission Critical

Systems on Hyper-V
Daniel Lai

(MCITP:Virtualization| MCSE:PrivateCloud | VCP:vSphere5.1 Datacenter Virtualization |

VCA:WorkforceMobility , Datacenter Virtualization, Cloud )

Solutions Sales Manager | WW Cloud Platform Tech Ranger |

Cloud Platform & Enterprise Mobility
Microsoft

VMware Comparison
System

Host

VM

Cluster
1.
2.

Windows Server
2012 R2 Hyper-V

vSphere
Hypervisor

vSphere 5.5
Enterprise Plus

Logical Processors

320

320

320

Physical Memory

4TB

4TB

4TB

Virtual CPUs per Host

2,048

4,096

4,096

Virtual CPUs per VM

64

641

1TB

1TB

1TB

1,024

512

512

Guest NUMA

Yes

Yes

Yes

Maximum Nodes

64

N/A2

32

8,000

N/A2

4,000

Resource

Memory per VM

Active VMs per Host

Maximum VMs

vSphere 5.5 Enterprise Plus is the only vSphere edition that supports 64 vCPUs. Enterprise edition supports 32 vCPU per VM with all other editions supporting 8 vCPUs per VM
For clustering/high availability, customers must purchase vSphere

vSphere Hypervisor / vSphere 5.x Ent+ Information: http://www.vmware.com/pdf/vsphere5/r55/vsphere-55-configuration-maximums.pdf and http://www.vmware.com/products/vsphere-hypervisor/faq.html,

http://www.vmware.com/files/pdf/vsphere/VMware-vSphere-Platform-Whats-New.pdf

Virtualized Workload Performance

Highest levels of performance
for key Microsoft workloads

1
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0

800
700
600
500
400
300
200
100

0
4

16

32

Virtual Processors Per VM

64

Average Transaction Response Time (Sec)

64 vCPU support drove 6x

performance increase over
previous version of Hyper-V
& 6.3% overhead compared
with physical equivalent

900

Transactions/Sec

SQL Server 2012

Hyper-V Virtual CPU Scalability

with OLTP Workloads

Virtualized Workload Performance

Virtualized 48,000 simulated

users on a single Hyper-V
host across 12 VMs, with low
response times

Exchange Workload Scalability on

Windows Server 2012 with Hyper-V
50000

50

45000

45

40000

40

35000

35

30000

30

25000

25

20000

20

15000

15

10000

10

5000

0
2

10

Virtual Machines
3 vCPU, 16GB RAM per VM, JetStress 2010

12

Exchange DB Read Response Time (ms)

Exchange 2013

Exchange 2013 Mailboxes

Highest levels of performance

for key Microsoft workloads

Virtualized Workload Performance

Scaled to over 2 million

heavy users at 1%
concurrency, across 5 VMs
on a single Hyper-V host

SharePoint Workload Scalability on

Windows Server 2012 Hyper-V
2,400,000
2,200,000
2,000,000
1,800,000
1,600,000
1,400,000
1,200,000
1,000,000
800,000
600,000
400,000
200,000
0

1
0.9
0.8

0.7
0.6
0.5
0.4

0.3
0.2
0.1
0

Web Front Ends

8 vCPU, 12GB RAM per WFE VM

Average Response Time (Sec)

SharePoint 2013

Heavy Users (1% Concurrency)

Highest levels of performance

for key Microsoft workloads

Virtual Machine Mobility

More flexible, higher performing VM migration capabilities

Live
Migration

SMB-Based
Live Migration

Live Migration
with Compression*

Faster, unrestricted,
simultaneous VM live
migrations between cluster
nodes with no downtime

Store VMs on SMB 3.0 file

shares, and live migrate
them flexibly, without the
need for a cluster

Harness spare host CPU

cycles to compress memory
of migrating virtual machines
to accelerate live migration

* New in Windows Server 2012 R2

Virtual Machine Mobility

More flexible, higher performing VM migration capabilities

Live Migration
over SMB*

Live Storage
Migration

Shared-Nothing
Live Migration

Cross-Version
Live Migration*

Highest live migration

performance, utilizing
SMB 3.0 protocol &
accelerated using RDMA
capable NICs

Move the virtual hard

disks of running virtual
machines to a different
storage location with
no downtime

Flexibly move virtual

machines between Hyper-V
hosts with nothing but a
network cable and no
clustering requirements

Migrate VMs from

Windows Server 2012
Hyper-V to 2012 R2
Hyper-V hosts with no
downtime

* New in Windows Server 2012 R2

High Availability & Resiliency

Provide the greatest levels of uptime for key applications

Scalable up to 64 Nodes & 8,000 VMs
Reduced Active Directory Dependencies*
VM Monitoring

Failover
Clustering
Integrated, scalable high
availability to protect key
virtualized workloads and
provide the highest levels
of uptime

Failover Priority, Affinity & Anti-Affinity

Cluster Shared Volumes Optimizations*
Cluster-Aware Updating

VM Network Health Detection*

Dynamic Witness & Force Quorum*

* New in Windows Server 2012 R2

High Availability & Resiliency

Provide the greatest levels of uptime for key applications

Failover
Clustering

Flexible Guest
Clustering

Integrated, scalable high iSCSI, Virtual Fibre

availability to protect key Channel & SMB 3.0
virtualized workloads andsupport for in-guest,
provide the highest levelsapplication level
of uptime
resiliency

Shared VHDX
Support*

Integrated
NIC Teaming

Build production-ready
guest clusters utilizing
Shared VHDX, without
exposing underlying
physical storage

Aggregate network
adaptors to increase
throughput & provide
redundancy in case of
link failure

* New in Windows Server 2012 R2

Backup & Replication

Integrated solutions for protecting key virtualized workloads

Incremental
Backup

Windows Azure
Backup Integration

Inbox VM
Replication

Perform agentless backup

operations more quickly &
easily whilst saving network
bandwidth & disk space

Quickly & easily protect

important server data offsite with
automated backups to Windows
Azure

Near-synchronous replication
of virtual machines across
multiple sites with Hyper-V
Replica

Cloud Witness Greater efficiency

Azure

Leveraging the power of the public cloud to

increase resiliency of your private cloud
Hybrid Cloud

Azure blob storage as an arbitration point

Stretched clusters without a 3rd site
Clusters without shared storage

Flexible Scenarios

Witness

Guest Clusters in Azure VM role

https
Cluster

Site1

Site2

VM Compute Resiliency Reliability

Designing for cloud scale with commodity
hardware
Flexibility

Configurable based on your SLAs

VMs continue to run even when a node falls out

of cluster membership
Availability

Resiliency to transient failures

Reliability

VMs continue
running when node
becomes isolated

Quarantine of Flapping Nodes Reliability

Unhealthy nodes are quarantined and are no longer
allowed to join the cluster
Protection

Prevents flapping nodes from negatively effecting other

nodes and the overall cluster
Node is quarantined if it ungracefully leaves the cluster
three times within an hour

Resiliency

VMs are gracefully drained once quarantined

QuarantineDuration: cluster property
Default 2 hours

Control

QuarantineState: Node read only common property

for node substatus

VM Storage Resiliency Reliability

Designing for cloud scale with standard hardware

Resiliency

Preserve tenant VM session state in the event of

transient storage disruption
VM stack quickly notified on failure

Visibility

Intelligent and quick VM response to block the

file based storage infrastructure issues
VM moved to PausedCritical state and will wait
for storage to recover

Reliability

Session state retained on recovery

Storage Replica More uptime

Cross site HA DR: Stretch clusters across sites with
synchronous volume replication
End-to-end Windows Server disaster recovery
solution
Integrated
management

Failover Cluster Manager UI and PowerShell

Works with any Windows volume, uses SMB3
as transport

Flexible

Hardware agnostic - works with Storage

Spaces or any SAN volume
Block-level synchronous volume replication

Scalable

Automatic cluster failover for low Recovery

Time Objective (RTO)

Multi-Site Cluster

Site1

Site2

Storage Spaces Direct Low cost

SoFS clusters with no shared storage. Doesnt need shared JBODs and SAS fabric behind
Scale Out File Server nodes

Cloud design
points and
management

Prescriptive configuration. Reduced hardware

costs with SATA drives

Deploy, manage and monitor with System Center

Hyper-V Cluster(s)

Fault tolerance to disk, enclosure, node failures

Reliability,
Scalability,
Flexibility

Scale pools to large number of drive

Fine-grained storage expansion

Hyper-V IaaS storage

Storage for Backup and Replication targets
Use Cases

SMB3 Storage Network Fabric

Hybrid Cloud Scenarios

Encrypted Backup
VPN
Windows Backup
SC Data Protection Manager

Recovery

Microsoft
Azure

System Center
Virtual Machine
Manager

Manage
Recovery
plan

Health Monitor
Microsoft
Azure

Site B

Replication

Site A

Recovery

Hyper-V
Replica

Site A

System Center
Virtual Machine
Manager

Microsoft
Azure

Site B

System Center
Virtual Machine
Manager

Orchestrated Recovery in case of outage

VPN

Admin

Remote Users

Hybrid Cloud Scenarios

File / Application
Servers

File / Application
Servers

Live Backups, Archives,

and Disaster Recovery
Dramatic Cost
Reduction
No Changes to
Application
Environment

File share with integrated

data protection
All-in-one primary data
+ backup + live archives
+ DR with de-duplication
& Compression

Most
Active
Data
on SSD

StoreSimple
Archive
Data
Encrypted

SharePoint storage on
StorSimple + Azure
StorSimple SharePoint
Database Optimizer
Improved
performance &
scalability

Storage for Tier 2 3

SQL Databases
Integrated Backup,
Restore & Disaster
Recovery

Warm
data on
SAS Local
Tier
Policies

De duplicated,
Compressed & Encrypted
De
duplicated

VPN
Encrypted Backup

De duplicated
& Compressed

Recovery

Automated

Control Virtual Sprawl

Cloud-as-a-tier
Offload storage footprint
VMware Storage DRS Storage
pools
Virtual Machine Archive
Regional VM Storage

Microsoft
Azure
Benefits
Consolidates primary, archive,
backup, DR thru seamless
integration with Azure
Cloud Snapshots
De duplication
Compression
Encryption
Reduces enterprise storage TCO
by 6080%

Hybrid Cloud Scenarios

Developers

VPN

Tier 1

Microsoft
Azure SDK

Tier 2

Availability Set

VPN

Microsoft
Azure
MICROSOFT
AZURE
CDN

User
s
On Premises

Microsof
t
Load
Auto Web Virtual VHD
Azure
Balancing Scaling Site Machines
AD
MICROSOFT
Storage
AZURE
BLOB
Cache
Storage
Queue

INGRESS NODES
Collect / Decode

Connected Devices

Microsoft
Azure

Load
Auto Worker
Balancing Scaling Roles

Tier 3

Availability Set

Availability Set

Auto SharePoint Mobile

Scaling
Service

Azure HDInsight
Analytics
SQL
& Reporting Azure Storage (Hadoop)

Storage
Table

ANALYTICS NODE
Record

Azure
Storage

Filter / Analyze / Aggregate

Auto Worker
Scaling Roles

Notification
Hub

CONSUME
Reporting / BI

Azure
Analytics
SQL
& Reporting Azure Storage

Azure
Management

System Center
foundation
Extension of todays
environment
Innovation through
cloud cadence

Starting the journey. Modern management

Containers
Wire Data
Configuration

Remote OS
Management

CMDB

Inventory

Linux
Patching

Alerting

Operations
Management Suite