Virtual Machine Mobility with Vmware VMotion

and Cisco Data Center Interconnect Technologies

What You Will Learn

VMware has been the industry leader in virtualization technologies for the past decade and has
brought to the data center several new features that enable faster and better provisioning of
® ™
business-critical applications. One of the features is the VMware VMotion technology that allows
™
virtual machine mobility between two VMware vSphere servers instantaneously with no application
downtime. The capability to migrate applications with no perceivable effect to the end user enables
IT departments to develop new and improved methods for provisioning and maintaining data center
infrastructure. IT departments can now perform hardware maintenance, consolidate CPU and
memory resources, or migrate mission-critical applications from a data center when necessary
without affecting the service-level agreements (SLAs) of the applications.

A successful application migration through VMware VMotion heavily relies on the underlying network
infrastructure. Therefore it is extremely important that IP network be resilient, robust, and highly
available.. The IP network becomes more important when the applications have to be mobile across
data centers. Cisco has been the industry leader in IP network and routing technologies and has
®
been providing data center IP network extensions since the 1980s. Cisco switching and routing
technologies provide the robust and redundant network that is essential for VMware VMotion to
succeed.

This document discusses the VMware VMotion feature and Cisco networking technologies essential
for application mobility across data centers.

VMware and Cisco Migration Solution

The VMware and Cisco solution shown in Figure 1 enables customers to perform live application
migration across data centers. The components used are a VMware vSphere 4.0 server cluster
enabled with VMware VMotion in each data center, a VMware vCenter server, and a data center
interconnect (DCI) WAN. The applications provisioned on the VMware vSphere server can be
migrated across the data centers or a private cloud with no application downtime. The solution itself
does not need any new software or hardware to perform these migrations.

© 2009 Cisco | VMware. All rights reserved. Page 1

 Figure 1. VMware VMotion Across Data Center

Need for Virtual Machine Mobility Across the Data Center

The changing model of data center management and provisioning allows VMware VMotion to be
used for several purposes without violating the application SLAs.

● Data center maintenance without downtime: Applications on a server or data center

infrastructure requiring maintenance can be migrated offsite without downtime.
● Disaster avoidance: Data centers in the path of natural calamities (such as hurricanes) can
proactively migrate the mission-critical application environment to another data center.
● Data center migration or consolidation: Migrate applications from one data center to
another without business downtime as part of a data center migration or consolidation effort.
● Data center expansion: Migrate virtual machines to a secondary data center as part of data
center expansion to address power, cooling, and space constraints in the primary data
center.
● Workload balancing across multiple sites: Migrate virtual machines between data centers
to provide compute power from data centers closer to the clients (―follow the sun‖) or to load-
balance across multiple sites. Enterprises with multiple sites can also conserve power and
reduce cooling costs by dynamically consolidating virtual machines into fewer data centers
(automated by VMware Dynamic Power Management [DPM]), another feature enabling the
green data center of the future.

The application mobility discussed in this document provides the foundation necessary to enable
cloud computing—for example, cloud import and export—providing the flexibility to move virtual
machines into the cloud from an enterprise data center, to move them between different clouds, and
to move them back into the enterprise data center.

© 2009 Cisco | VMware. All rights reserved. Page 2

 VMware VMotion Requirements
VMware VMotion application mobility is based on certain infrastructure requirements:

● An IP network with a minimum bandwidth of 622 Mbps is required.

● The maximum latency between the two VMware vSphere servers cannot exceed 5
milliseconds (ms).
● The source and destination VMware ESX servers must have a private VMware VMotion
network on the same IP subnet and broadcast domain.
● The IP subnet on which the virtual machine resides must be accessible from both the source
and destination VMware ESX servers. This requirement is very important because a virtual
machine retains its IP address when it moves to the destination VMware ESX server to help
ensure that its communication with the outside world (for example, with TCP clients)
continues smoothly after the move.
● The data storage location including the boot device used by the virtual machine must be
active and accessible by both the source and destination VMware ESX servers at all times.
● Access from VMware vCenter, the VMware Virtual Infrastructure (VI) management GUI, to
both the VMware ESX servers must be available to accomplish the migration.

Challenges
Figure 2. Infrastructure Challenges to Vmware Vmotion Across Data Centers

Facilitating VMware VMotion migration across physical infrastructure boundaries and data centers
using the WAN poses specific challenges for the data center network and storage designs as shown
in Figure 2. In particular, the LAN and SAN designs have to be addressed.

● LAN (Layer 2 Domain Elasticity): The LAN must be extended across the physical locations
or data centers without compromising the availability, resilience, and security that exists
within a single physical location or data center. The best practices for the current network

© 2009 Cisco | VMware. All rights reserved. Page 3

 deployment require separate virtual LANs (VLANs) for the different physical locations. The
extension of the VLAN across data centers must be implemented in a way that preserves the
configuration and security constraints of a single data center.
● Data availability (Storage Elasticity): The availability of data to the two VMware ESX
servers is critical to a successful migration of the application. The SAN design should
encompass all the properties and parameters of the data within the same data center. This
requirement poses a heightened security risk, since data now has to traverse the WAN. Data
network designs should take into account these parameters to help ensure that data is not
only available, but secured and available with I/O latencies that will not affect the SLAs of the
applications.
● IP localization: A primary requirement for application mobility is that the migrated virtual
machine maintains all its existing network connections after it has been moved to the
secondary data center. Traffic routing to and from the virtual machine needs to be optimized
so that any traffic flows in an optimized way to the virtual machine’s new location. If the traffic
to the virtual machine originates in the same Layer 2 domain, then the Layer 2 extension will
suffice. However, if the traffic to the virtual machine is traversing a Layer 3 network or the
Internet, then granular routes needs to advertised by the secondary data center for migrated
virtual machines. If these changes are not provisioned, suboptimal routing may result in
additional delay, which may or may not be acceptable depending on the specific virtual
machine or application. To help ensure that the traffic from the virtual machine is optimally
routed, the default gateway IP addresses on the subnets in both the primary and secondary
data centers need to be identical, a goal achieved through active-active Hot Standby Router
Protocol (HSRP) configuration.
● Services localization: Another critical requirement is helping ensure that any applicable
services (firewall, load balancer, etc.) policies are retained as the virtual machine moves from
one data center to another data center.

In addition, WAN characteristics such as available bandwidth for VMware VMotion and SAN
extension and its latency are crucial factors in the eventual success of application mobility.

Solution Options
The challenges that arise from migrating applications across data centers are addressed by the joint
VMware and Cisco solution. The solution uses best practices for the LAN, data network, and WAN to
overcome the challenges and provide the IT staff with a tool for migrating applications between data
centers without downtime.

The solution can be designed in multiple ways based on the topologies used for the LAN and
storage extensions, as shown in Table 1.

© 2009 Cisco | VMware. All rights reserved. Page 4

 Table 1. Application Mobility Using Vmware Vmotion Solution Options

Network and Shared Storage Active-Passive Active-Active

Storage Storage Storage
Topologies

Extended or Storage remains at Storage is migrated Storage is actively

Stretched VLAN original location before virtual machine available at both
migrates locations

Extended VLAN with Shared Storage

An extended VLAN and shared storage architecture extends the VLAN between the two sites, but
with storage remaining at the original location. When the virtual machine migrates to the remote data
center, the application will access the storage from the primary site. Storage is not provisioned for
the application at the remote data center; hence, there is only one copy of the storage at any given
point in time. This design can be appropriate when the distance between the data centers is not
great, since I/O latency will affect application performance.

Extended VLAN with Active-Passive Storage

An extended VLAN with VMware Storage VMotion topology requires migration of storage to the
remote data center prior to migration of the virtual machine itself to the remote data center. Storage
is migrated to remote data center prior to the virtual machine migration using VMware Storage
VMotion. VMware Storage VMotion migrates the data space associated with a virtual machine to the
secondary storage location and enables a virtual machine to access this new storage after the
VMware Storage VMotion migration is complete.

If storage replication were in place between the data centers, the volume(s) containing the virtual
machine data could be readily available in real time at the secondary data center. The existing
Active-Passive storage replication techniques require a set of explicit control operations to make the
storage replica accessible to the servers in the secondary data center. Consequently, at present, this
is not a supported technology to perform virtual machine vMotion.

Extended VLAN and Active-Active Storage

An extended VLAN and active-active storage solution incorporates technologies that make data
actively available at both the local and remote data centers at all times. The LAN extends across the
data centers, and storage is provisioned in both data centers. Data is replicated across data centers
using synchronous replication technology and rendered in an active-active state by the storage
manufacturer. Normally when data is replicated, the secondary storage is locked by the replication
process and is available to the remote server only in a read-only state. In contrast, active-active
storage allows both servers to mount the data with read and write permissions as dictated by the
VMware VMotion requirements.

VMware VMotion Across Very Long Distances (Routed or Disparate IP Subnets)

Deploying VMware VMotion across data centers that are dispersed over very long distances (500
miles or more) potentially involves moving the virtual machine to an entirely new subnet, but the goal
continues to be to help ensure that the IP address of the virtual machine as well as the existing client
connections are not disrupted. This type of VMware VMotion migration is not possible with existing
technologies. Special hardware and software features will be required to route the TCP connections

© 2009 Cisco | VMware. All rights reserved. Page 5

 to the virtual machine in its new location without terminating the sessions. This approach will require
the redesign of the IP network between the data centers involving the Internet. Technologies are
being developed by Cisco, VMware, and standards organizations to address this network scenario in
the future.

VMware and Cisco Solution

The VMware and Cisco solution, jointly engineered by the two companies, addresses two of the
available topologies: extended VLAN with shared storage and extended VLAN with active-passive
storage. The two components of this solution are LAN extension technologies and SAN extension or
storage availability technologies.

LAN Extension Technologies

To make the same LAN to be available across the data center, Cisco has solutions to suit the type of
DCI available in your network. Any solution that extends the Layer 2 subnet across data centers
needs to meet the following requirements:

● High availability: The solution must help ensure that no link or device failure will cause the
Layer 2 extension to be disconnected and thus affect the ability to perform VMware VMotion
migration across the data center.
● Load balancing: The solution should fully utilize cross-sectional bandwidth between the data
centers; DCI connections are usually more expensive than the LAN, making effective use of
the available connection critical.
● Spanning Tree Protocol isolation and loop and broadcast storm prevention: The
solution must fully contain and isolate Spanning Tree Protocol within each data center with
Bridge Protocol data units (BPDUs) filtered at the boundary of each edge switch facing the
core. Network loop and broadcast storm avoidance features need to be available to prevent
disruption of applications.
● Scalability: The solution must be able to handle multiple VMware VMotion migrations
concurrently. To meet this requirement, the network must be able to scale. The available
bandwidth, the number of VLANs, and the number of data centers connected through the
solution should all be capable of expansion as needed.

Additional services may be required in many cases:

● Encryption: IP Security (IPsec) or Layer 2 link encryption (IEEE 802.1AE) may be needed to
help ensure the privacy and confidentiality of the data traversing between the data centers.
This requirement is particularly important if the interconnection crosses a public network.
Encryption may also be a requirement for regulatory compliance.
● Hierarchical quality of service (HQoS): HQoS may be required to help ensure quality of
service (QoS) for VMware VMotion, particularly on WAN links acquired through a service
provider. HQoS is important on interconnecting devices when an enterprise subscribes to a
subrate service provider service.

© 2009 Cisco | VMware. All rights reserved. Page 6

 Table 2 lists the LAN extension options for each type of DCI.

Table 2. Cisco LAN Extension Solutions and Platforms for Various Transport Options

Type of Cisco LAN Extension Cisco Platform

Interconnect Solutions
(Transport
Option)
®
Dark Fiber or Dense Virtual switching system (VSS) Cisco Catalyst 6500 Series Switches
Wave-Division
Multiplexing
(DWDM) ™
Virtual PortChannel (vPC) Cisco Nexus 7000 Series Switches

Crossponder Cisco ONS 15454 crossponder

Multiprotocol Label Ethernet over MPLS (two data ● Cisco Catalyst 6500 Series Shared
Switching (MPLS) centers) Port Adapter (SPA) Interface
Processor 400 (SIP-400) and SIP-
600
● Cisco ASR 1000 Series
Aggregation Services Routers

Virtual Private LAN Service ● Cisco Catalyst 6500 Series SIP-

(VPLS) (multiple data centers) 400 and SIP-600

IP Ethernet over MPLS (EoMPLS) ● Cisco Catalyst 6500 Series SIP-

over Generic Routing 400
Encapsulation (GRE) (2 data ● Cisco ASR 1000 Series
centers)

VPLS over GRE (multiple data ● Cisco Catalyst 6500 Series SIP-
centers) 400

Additional services such as encryption and HQoS can be implemented on the Cisco hardware listed
in Table 3.

Table 3. Cisco Solutions and Platforms for Additional Services for LAN Extension Schedules

Service Solution Platform

Encryption IEEE 802.1AE Cisco Nexus 7000 Series

IPsec ● Cisco Catalyst 6500 Series SPA Services Card 600

(SSC-600) and VPN Services Port Adapter (VSPA)
● Cisco ASR 1000 Series

Multilevel QoS HQoS ● Cisco Catalyst 6500 Series SIP-400 and SIP-600
● Cisco ASR 1000 Series

© 2009 Cisco | VMware. All rights reserved. Page 7

 For more information about Cisco LAN extension solutions for data center interconnection, refer to
http://www.cisco.com/en/US/netsol/ns975/index.html.

Storage Extension Technologies

The availability, scalability, security, and performance of the storage subsystem are of utmost
importance to any enterprise. The task of ensuring that all these factors are addressed in a single
data center is a daunting task for any storage administrator; extending them across data centers is
an even greater challenge, requiring implementation of storage best practices. These factors directly
affect application performance, in turn affecting the SLAs of business-critical applications. The Cisco
MDS 9000 Family of SAN switches is especially suited to these SAN topologies. Table 4
summarizes the features that can be used to address the requirements for storage across data
centers.

Table 4. Cisco SAN Extension Solutions

Feature Requirements Functions

Virtual SAN Isolation and security The VSAN technology provides secure hardware-based
network segmentation, similar to the VLAN technology that is
widely deployed in LANs. Fabric services such as zoning and
routing are independent per VSAN.

In this validated solution, the nodes in each VMware ESX

cluster are placed in a dedicated VSAN, to use a consolidated
physical infrastructure and to be isolated with respect to
security threat and fabricwide errors.

Management and Cisco MDS 9000 NX-OS Software management offers several
access control levels of role-based access control (RBAC). This feature
allows an administrator to be in charge of a specific VSAN
without having any visibility into other VSANs.

The administrator can map the roles defined in the VMware

vCenter; for instance, an administrator may be able to access
a specific VSAN and the corresponding VMware ESX cluster
and nothing else.

Inter-VSAN Isolation and security In a DCI solution, each data center can implement
Routing independent VSANs, preserving the fabric services
(IVR) segmentation, data isolation, and administration
independence. IVR allows selected devices from different
VSANs, even across different data centers, to communicate
without any fabric merging.

In this validated solution IVR provides connectivity between

the VMware ESX servers located in the secondary data center
and the storage located in the primary data center (shared
storage). IVR can also provide connectivity to execute
VMware Storage VMotion across data centers and to perform
primary-array-to-secondary-array storage replication.

SAN Integrated solution The capability to plug long-wave and Coarse Wavelength

© 2009 Cisco | VMware. All rights reserved. Page 8

 Feature Requirements Functions
extension Division Multiplexing (CWDM) optics into the Cisco MDS 9000
with dark Series Switches simplifies SAN extension over dark fiber. The
fiber performances are guaranteed by the extended buffer-to-buffer
credits available with the Cisco MDS 9000 Series.

Security Cisco MDS 9000 Series Switches provide Cisco TrustSec

Fibre Channel Link Encryption to secure SAN extension data
across native Fibre Channel links

SAN Integrated solution Cisco MDS 9000 Series Switches provide Gigabit Ethernet
extension interfaces and support the FCIP protocol, to transparently
with FCIP extend the SAN over an IP network.

Security The Cisco MDS 9000 Series provides native IP Security

(IPsec) encryption to secure FCIP links.

Port Availability Cisco MDS 9000 Series PortChannels are the aggregation of
channeling multiple physical Fibre Channel or FCIP links into one logical
link, to provide higher aggregated bandwidth, load balancing,
and link redundancy.

I/O Application IOA is an intelligent distributed fabric service built into Cisco
acceleration performances MDS 9000 Series Switches. IOA accelerates I/O performance
(IOA) across distances. This feature helps the overall application
performance remain relatively the same, even when the
application server and the storage are separated by
considerable distance. In this validated solution, I/O
performance has been enhanced over the FCIP link.

All the features listed in Table 4 make a Cisco MDS 9000 Family SAN resilient and highly available.
More information about the Cisco MDS 9000 Series Switches can be obtained from
http://www.cisco.com/en/US/products/hw/ps4159/ps4358/prod_white_papers_list.html.

Solution Reference Architecture and Validation

To prove the validity of the solution, VMware and Cisco configured the solution as shown in Figure 3,
simulating the WAN and migrating a live application across the data centers without any
downtime.The configurations of the VMware vSphere server, the LAN within the data centers, the
SAN, and the WAN have been designed to enable VMware VMotion across data centers while
adhering to the VMware VMotion requirements.

The solutions described here have been jointly validated in the VMware and Cisco Joint Solutions
Lab. The validated topology used for the testing is shown in Figure 3.

© 2009 Cisco | VMware. All rights reserved. Page 9

 Figure 3. Jointly Validated Architecture

The network topology used in the joint solution test simulates two data centers extended over
different types of DCIs. Data center 1 is a Cisco best-practices three-tier architecture with Cisco
Catalyst 6500 Series Switches forming the core and aggregation layers and a pair of Cisco Catalyst
4900M Switches forming the access layer. VSS technology is used to provide the extended (or
stretched) VLAN. Data center 2 is also designed using the Cisco three-tier architecture, but with the
Cisco Nexus 7000 Series Switches and the Nexus 5000 Series Switches. The functional architecture
at both data centers is similar, with VSS and vPCs used to provide the extended VLAN.

The storage for the solution is provisioned using either of two methods depending on the test being
performed:

● Shared storage: Storage is located in data center 1, and the SAN is extended to data center
2 using FCIP SAN extension. FCIP IOA is enabled to help ensure that the application
performance does not suffer when the application is accessing its storage across the DCI
switches.
● Active-passive storage using VMware Storage VMotion: Storage is provisioned at both
locations. The storage capacity is provisioned identically at both data centers, and the
storage is presented to the VMware ESX servers as unique data stores. VMware Storage
VMotion migration is performed from one data store to the other; later, the VMware VMotion
virtual machine migration is performed. The storage must be available at all times to both the
source and destination VMware vSphere servers for VMware VMotion migration to be
successful. The SAN extension method used is again FCIP with IOA enabled.

© 2009 Cisco | VMware. All rights reserved. Page 10

 Test Topology and Tools
The solution was validated using real-life application servers migrating across data centers while
clients were accessing these applications. The applications used were Microsoft SQL Server and
Microsoft Exchange Server. Table 5 lists the configurations and the test tools used.

Table 5. Cisco and Vmware Vmotion Migration Across Data Center: Test Topology Summary

Application Server Stress- Application Description

Configuration Generation Performance
Tool Metrics

Microsoft SQL CPU: 4 virtual Dell DVD Store Orders per minute The DVD Store
Server 2005 CPUs (vCPUs) open source (OPM) benchmark is an
(64-bit) benchmark online transaction
Memory: 8 GB
processing
Storage: (OLTP)
Approximately 300 benchmark that
GB (EMC simulates the
CLARiiON) operation of a
DVD store.
OS: Microsoft Performance is
2008 64-bit server measured in
OPMs, indicating
the number of
orders
successfully
inserted into the
database per
minute.

Microsoft CPU: 4 vCPUs Microsoft Mail messages LoadGen is a

Exchange LoadGen being handled by Microsoft tool
Server 2007 Memory: 8 GB Microsoft developed to
Storage: Exchange Server closely mimic the
Approximately 300 operation
GB (EMC performed by
Symmetrix V-Max) Microsoft Outlook
clients on a
OS: Microsoft Microsoft
2003 32-bit server Exchange Server.
LoadGen is a
standard test tool
used to perform
Microsoft
Exchange Server
load and stress
tests.

Test Methodology

Microsoft SQL Server Test

● Reinitialize the Microsoft SQL Server by rebooting the VMware ESX server on which it is
resident and the target VMware ESX server to reset the statistics data.
● Start the DVD Store client on a virtual machine that has IP connectivity to both VMware ESX
servers.

© 2009 Cisco | VMware. All rights reserved. Page 11

 ● Run the DVD client and wait for 30 minutes for the client to attain a steady state; note the
OPM on that VMware ESX server.
● Migrate the system to the corresponding target.
● Wait for 30 minutes for the client to attain steady state; note the OPM for that VMware ESX
server.
● Perform 18 more migrations with a 10-minute wait between each migration.
● Collect test statistics to evaluate the total elapsed time.

Data Center Evacuation Test

1. Start the LoadGen and DVD Store clients on the respective client virtual machines.
2. Start the migration of the application servers in a sequential manner.
3. Collect test statistics to evaluate the total time required to migrate the data center.
4. Migrate the application servers simultaneously.
5. Collect statistics to evaluate the total elapsed time.

Test Results
The goal of the joint testing is to measure the duration of the overall VMware VMotion migration
process during VMware VMotion migration. The tests measure the time taken for the overall VMware
VMotion migration to be complete, beginning from the initiation of the VMware VMotion process. The
overall migration time is an important measure, and it becomes critical when multiple VMware
VMotion migrations are being performed. The amount of time for the overall migration depends on
the duration of each VMware VMotion migration. The duration of a VMware VMotion migration
largely depends on the distance between the source and destination VMware ESX servers and the
amount of bandwidth available between the data centers.

The application used to validate the solution is an e-commerce suite hosted on a Microsoft SQL
Server 2005. DVD Store Version 2 (DS2) is a complete online e-commerce test application with a
back-end database component, a web application layer, and driver programs. The virtual machine
hosting the back-end Microsoft SQL Server database is migrated across the data centers, and the
performance of the application in OPMs is captured.

Figure 4 shows the VMware VMotion migration times as a virtual machine is migrated from one
VMware ESX server to another, with the servers separated by different distances. The elapsed time
increases with distance, but it is a directly related to the network latency and the amount of network
bandwidth available for VMware VMotion as the distance increases. In the test scenario, the
application client maintained all the sessions, and a momentary drop in performance was observed
before the performance returned to steady-state values.

© 2009 Cisco | VMware. All rights reserved. Page 12

 Figure 4. Duration of Intra– and Inter–Data Center Vmware Vmotion Migration

The performance of the application when the back-end database is moved away from the users to a
data center 200 km away is shown in Figure 5. The results show that the migration to a remote data
center is feasible not only in terms of the mobility of the application but also from a business
perspective since degradation is less than 15 percent. Shared storage for the application therefore is
a viable option; with advanced SAN extension technologies such as FCIP and IOA, application
performance is maintained well within the SLA of the application.

The performance graph shows the performance of the application within the local data center, in
which the storage is located. The graph then shows the performance of the application with
increasing distance as it migrates to the remote data center, demonstrating that performance is
within the range of the SLA, which makes the SAN topology of shared storage for application
workload balancing a viable option.

© 2009 Cisco | VMware. All rights reserved. Page 13

 Figure 5. Microsoft SQL Server DVD Store Performance

In addition to migrating one mission-critical application across the distance of 200 km, the joint
testing migrated more applications in the data center to validate evacuation of the data center for
disaster avoidance (Figure 6). The joint testing included a Microsoft Exchange Server 2007 with
1000 Microsoft Outlook 2007 LoadGen users. The application performance monitored by LoadGen is
shown in Figure 7. The Microsoft Exchange Server is a four-vCPU virtual machine with 8 GB of
memory; resource utilization was 80 to 90 percent for the CPU, with approximately 20 Mbps of data
being read and written to the disks. Simultaneously, the Microsoft SQL Server running the DVD
Store database was also migrated. The OPM values for the DVD Store application are shown in
Figure 5. Figure 6 shows the migration times when the two applications are migrated simultaneously.
The results clearly show that the elapsed time for the Microsoft SQL Server increases by a small
amount, which is acceptable since the VMware VMotion network is now being shared with an
extremely busy Microsoft Exchange Server workload. In spite of this increase, there was no
perceivable effect on the clients performing the benchmark.

© 2009 Cisco | VMware. All rights reserved. Page 14

 Figure 6. Migration Times with Multiple-Application VMware VMotion Migration

Figure 7. LoadGen Results Summary

© 2009 Cisco | VMware. All rights reserved. Page 15

 Recommended Operational Procedure
The recommended procedure for implementing the joint VMware and Cisco solution is to have the
VMware vSphere high-availability clusters independent of each other in the two data centers.
VMware VMotion migration across data centers should be a manually instantiated task to keep
VMware Dynamic Resource Scheduling (DRS) from automatically moving virtual machines across
data centers.

Conclusion
VMware VMotion enables data centers to transparently implement virtual machine mobility using the
Cisco LAN and storage extension solutions. The VMware vSphere Virtual Data Center Operating
System (vDCOS) with the suite of features bundled with vSphere allows customers to transparently
migrate or evacuate data center applications with no downtime from a user perspective. This
technological capability gives IT departments tools to redefine the business continuance and disaster
recovery plans of the enterprise. The need for a more complex and expensive solution to meet the
recovery-point objective (RPO) of the business continuance plan can be reduced with the use of
VMware VMotion. The optimal architecture of the underlying transport infrastructure—the IP network
and SAN—enhances the solution to bring the RPO to near zero, and also reduces the recovery time
objective (RTO) to a very small number. The joint Cisco and VMware solution gives IT departments a
very powerful tool for better provisioning, utilizing, and maintaining a virtualized data center with
resources spread across multiple physical locations.

For More Information

● VMware VMotion: http://www.vmware.com/products/vi/vc/vmotion.html
● Data Center Interconnect (DCI): Layer 2 Extension Between Remote Data Centers:
http://www.cisco.com/en/US/prod/collateral/switches/ps5718/ps708/white_paper_c11_49371
8.html
● Cisco Catalyst 6500 Series Switches: http://www.cisco.com/go/6500
● Cisco Nexus 7000 Series Switches: http://www.cisco.com/go/nexus7000
● Cisco MDS 9000 Family: http://www.cisco.com/go/mds

© 2009 Cisco | VMware. All rights reserved. Page 16

 Cisco Systems, Inc. VMware, Inc
170 West Tasman Drive 3401 Hillview Ave
San Jose, CA 95134-1706 Palo Alto, CA 94304
USA USA
www.cisco.com www.vmware.com
Tel: 408 526-4000 Tel: 1-877-486-9273 or 650-427-5000
800 553-NETS (6387) Fax: 650-427-5001
Fax: 408 527-0883

Copyright © 2009. VMware, Inc. All rights reserved. Protected by one or more U.S. Patent Nos. 6,397,242, 6,496,847, 6,704,925, 6,711,672, 6,725,289, 6,735,601, 6,785,886, 6,789,156, 6,795,966,
6,880,022, 6,944,699, 6,961,806, 6,961,941, 7,069,413, 7,082,598, 7,089,377, 7,111,086, 7,111,145, 7,117,481, 7,149, 843, 7,155,558, 7,222,221, 7,260,815, 7,260,820, 7,269,683, 7,275,136,
7,277,998,7,277,999, 7,278,030, 7,281,102, 7,290,253, 7,356,679 and patents pending.
Cisco, the Cisco logo, and Cisco Systems are registered trademarks or trademarks of Cisco Systems, Inc. and/or its affiliates in the United States and certain other countries. All other trademarks
mentioned in this document or Website are the property of their respective owners. The use of the word partner does not imply a partnership relationship between Cisco and any other company.

(0812R) C11-557822-00 08/09

© 2009 Cisco | VMware. All rights reserved. Page 17