Professional Documents
Culture Documents
Using HP 3par Peer Persistence To Achieve Non-Disruptive High Availability and Load Balancing in Vmware Vsphere Environments
Using HP 3par Peer Persistence To Achieve Non-Disruptive High Availability and Load Balancing in Vmware Vsphere Environments
High availability for all virtualized applications locally or in the cloud is the aspiration of almost every enterprise
regardless of its size. It is time to make that dream a reality. Using HP 3PAR StoreServ, its Peer Persistence
software and VMware vSphere vMotion and Metro Storage Clusters, enterprises are closer than ever to delivering
on this higher form of high availability.
Enterprises of all sizes are coming to the realization that they should focus on uptime’s
value as opposed to downtime’s costs. As this transition in thinking occurs, applying high
availability (HA) and application load balancing features for their virtualized environments
extend beyond their primary data center to other data centers and even the cloud. The HP
3PAR StoreServ with its Peer Persistence feature gives enterprises access to these HA
and load balancing features they seek so they may keep their VMware environment online
Company all the time while better utilizing their available physical resources.
Hewlett-Packard Company
3000 Hanover Street BCP is not HA
Palo Alto CA 94304-1185
(650)857-1501 Business continuity planning (BCP) is already a mainstay at many midsized enterprises. An
AT&T 2012 online survey of IT executives from organizations throughout the United States
Founded 1939
www.hp.com found that over 80% of them, to include businesses with revenues of no more than
Industry $25 million, have BCP in place.1 These survey respondents who have the primary respon-
Computer Systems sibility for BCP in their organizations also shared the following:
Compounding the problem, enterprises are questioning BCP’s overall value. Aside from
the cost to build or lease a secondary site to host these BCP exercises, the computer gear
June 2013
Executive White Paper > Hewlett-Packard
in these sites often sits idle or heavily underutilized as it is only used during BCP activities while consuming
operating expenses the rest of the time.
The shortcomings of BCP explain why enterprises are coming to the conclusion that uninterrupted applica-
tion availability with its automated application failover and failback is a better option than BCP. The ready
availability of cloud technologies means that almost any size enterprise can realistically implement and main-
tain this environment—potentially more affordably and easily than BCP. Yet to achieve this higher form of HA
for all of their applications requires they first put in place the right set of technologies.
2. Array-based synchronous replication. VMware vMotion and MSC moves VMs back and forth
between different physical machines at different sites. However the data of these VMs must be
moved separately. To ensure the data is at the target site as a VM comes up requires the use of
array-based synchronous replication software such as HP 3PAR Remote Copy which keeps the
data associated with the VMs in sync between the two sites.
3. ALUA. Asymmetric Logical Unit Access (ALUA) is used by hypervisors such as VMware vSphere
to communicate with backend storage arrays. ALUA provides multi-pathing (two or more storage
networking paths) to the same LUN on a storage array and marks one path “Active” and the
other “Passive.” The status of the paths may be changed either manually by the user or program-
matically by the array.
4. “ALUA” aware storage arrays. Should the status of paths be reversed (“Active” paths become
“Passive” and vice-versa), the storage array must notify the hypervisor host appropriately so that
the hypervisor can re-scan the network for currently active paths.
6. Quorum Witness. A Quorum Witness is software that resides on either a physical or virtual
machine and monitors the availability of each site. Should one site go down or offline, it detects
the status of the failed site, notifies both VMware and the HP 3PAR StoreServ array of this
condition and has both of them failover application operations to the alternate site.
Two HP 3PAR StoreServ storage arrays that span two locations are presented to a VMware vSphere Metro
Storage Cluster (MSC) as a single stretched cluster using its Peer Persistence Software. The HP 3PAR
StoreServ arrays use Remote Copy to synchronously replicate data to keep all of the data on the LUNs
presented to the VMs in the vSphere MSC in sync with one another.
2 June 2013
Since HP 3PAR StoreServ controllers are Active-Active, the vSphere cluster at the primary production does
not need to dedicate ALUA to doing path management at the primary site. Rather the vSphere MSC uses
ALUA at the secondary site and marks its path to the HP 3PAR StoreServ array at this location as “Passive.”
Now the HP 3PAR Peer Persistence software plays its part. The HP 3PAR Peer Persistence software and
vSphere work together to detect any path failures for any VM. Should vSphere detect a path failure at the
primary site, vSphere will activate the “Passive” path at the secondary site and may vMotion the VM to a
physical server at the secondary site.
vSphere will at the same time generate an alert that the path at the secondary site is “Active.” A storage
administrator may then leverage the HP 3PAR StoreServ Peer Persistence software to issue a switchover
command to change the status of the “Passive” path at the failover site to “Active” and downgrade the
“Active” path at the primary site to “Passive.”
On the array side, the Quorum Witness detects the failure and notifies the HP 3PAR Peer Persistence soft-
ware to mark all paths at the recovery site as “Active” so regardless of what action VMware MSC takes (path
redirect or complete failover of all VMs) application processing may continue uninterrupted. The Quorum
Witness triggers this recovery without any manual intervention to ensure continuous application availability
with no VM restarts.
Unlike in the past when enterprises built or leased recovery sites and then saw these resources either sit
idle or go largely unused, they can now unlock their value in a number of ways. Minimally they can create
private clouds with production VMs hosted at both the primary and secondary sites. In this way all resources
are used and should perform better while still giving them the flexibility to automatically fail all of the VMs to
the other site should some type of disaster occur.
The other intriguing possibility is that enterprises may even have the option to lease server and storage
capacity from a cloud provider. Three out of four of the world’s largest service providers already use HP 3PAR
StoreServ as their storage solution. By selecting one of these providers as their cloud partner, enterprises
Bottom line: Some midrange storage arrays offer some of these features found on the HP 3PAR
StoreServ. Only HP 3PAR StoreServ offers all of the features on a storage array that are needed to
deliver this higher form of HA that enterprises now seek.
Using the HP 3PAR StoreServ and its Peer Persistence feature in conjunction with VMware vSphere,
enterprises can now do what has been largely impossible. It puts these higher forms HA and load
balancing within reach for VMware vSphere-hosted apps that other storage solutions at the mid-tier
level simply cannot match.
1. AT&T. 2012 AT&T Business Continuity Study. Rep. AT&T, May 2012. Web. 22 May 2013.
http://www.att.com/Common/about_us/pdf/national_business_continuity_summary_2012.pdf
2. Ibid.
©2013 DCIG, LLC. All rights reserved. All other brands or products are trademarks or registered trademarks of their respective holders and should be treated as such. The information, product recommendations and opinions made by DCIG LLC are
based upon public information and from sources that DCIG LLC believes to be accurate and reliable. However since market conditions change, the information and recommendations are made without warranty of any kind. All product names used and
mentioned herein are the trademarks of their respective owners. DCIG LLC assumes no responsibility or liability for any damages whatsoever (including incidental, consequential or otherwise) caused by one’s use or reliance of this information or the
recommendations presented or for any inadvertent errors which this document may contain. Any questions please call DCIG LLC at (402)884-9594.
4AA4-6976ENW, Created June 2013