Close this search box.

New Primary Data Deduplication Feature in Symantec FileStore N8300 v5.7 Targets Today’s VMware Deployments

VMware will hit a tipping point with 2011 with VMware customers expected to cross the 50% virtualization threshold by the end of the year. But as VMware adoption accelerates, a hurdle that every organization faces as it looks to implement VMware vSphere is identifying a back-end storage system that delivers the appropriate levels of availability and performance at the right price point. Last week the new Symantec FileStore N8300 v5.7 with its new primary data deduplication feature gave organizations a glimpse into how that may be accomplished.

The statistics surrounding the adoption rate and growth of server virtualization in general and VMware specifically over the last few years are staggering. Consider:

  • Over 19,000 customers, partners, press and analysts attended the 2011 VMworld conference in Las Vegas
  • A new VM is created somewhere in the world every 6 seconds
  • 10 million VMs existed at the end of 2010

But that does not explain why even now in late 2011 organizations are not realizing the savings that they expected server virtualization to deliver. Because even as they reduce their server costs, their storage costs – both capital and operational – are continuing to rise as they deploy external, network attached storage systems to support their growing virtualized environments.

Standing in the way of them lowering their storage costs is the lack of cost-effective, highly available, storage systems that deliver the performance and scalability that VMware requires. 

The storage systems that most enterprises use today were designed well before server virtualization exploded onto the scene. So while many of these storage systems have been adapted to work with VMware, they still do not address one or more fundamental issues such as:

  • Cost too much on a per GB basis
  • Do too little to optimize storage capacity (i.e. – no primary data deduplication features)
  • Do not put virtualized data on the right tier of storage at the right time
  • Lack necessary availability and reliability features
  • Cannot scale easily performance and capacity
  • Too difficult to manage

In short, what VMware environments need is an enterprise caliber storage system whose capital and operational costs do not offset the savings that server virtualization offers. Last week’s v5.7 release of the Symantec FileStore N8300 provides the additional features that storage system need to offer so organizations can begin to realize the broader cost savings that they expect VMware to deliver

Three features as specifically related to VMware that stand out in the Symantec FileStore N8300 v5.7 release include:

  • The introduction of primary block-level data deduplication. High levels of redundant data on VMware virtual machines (VMs) make block-level data deduplication almost a necessity in order for organizations to effectively control storage growth in these environments.  However it is not so much deduplicating existing data where organizations will see the most benefit from the N8300’s new data deduplication feature. Rather it is controlling storage growth as they create new VMs.

One of the challenges that every organization faces with VMware is that VMware vSphere makes it very easy to create new VMs for new applications. The downside of this ease of growing one’s application environment is that each new VM requires additional storage capacity. Allocating this storage capacity for each VM now begins to eat into the savings that VMware initially provided.

This is where the N8300’s new primary data deduplication feature should provide the greatest savings. Rather than allocating more storage for each new VM as it is created, the N8300 creates a clone or snapshot of a primary or “golden image” VM with the new VM pointing to the original image. Since each new VM initially only requires minimal additional storage capacity (usually a few GBs at most,) organizations can grow their VMware farm without running up their storage costs.

The FileStore N8300 v5.7 also addresses performance concerns that organizations may have about deduplicating the data on their VMs. The N8300 retains frequently accessed data in its read cache. Since the most frequently accessed files (such as operating system files) on VMs are almost always the same, it is highly probable that by the FileStore N8300 v5.7 first deduplicating and then storing frequently accessed VM data in its read cache the boot performance of VMs will show sign significant performance gains even as storage costs are reduced.

  • FileStore N8300 v5.7 integrates with vCenter and is VMware certified. Being on the VMware hardware certified list (HCL) is a prerequisite for any storage system to even get in the door in today’s enterprise data centers. This certification the FileStore N8300 v5.7 provides day one. But just as important as the VMware certification is integration with vCenter to effectively manage the attached storage array from vCenter.

As more organizations consolidate their data centers, server and storage administrators are transitioning to become “infrastructure administrators.” In this role they need to manage their entire data center from a single management console which, for many, is becoming vCenter.

The FileStore N8300 v5.7 provides this needed vCenter integration out of the box. However it does so in such a way that as each VM is created through vCenter, it capitalizes on the new data deduplication features by only allocating the additional amount of storage capacity that each VM needs to store the data that is unique to it.

This eliminates the need for administrators to first create the VM using vCenter and then login onto FileStore to allocate storage to it. Instead this VM creation and storage allocation process may now occur as a single transaction executed within vCenter. While administrators will first need to setup policies within vCenter to automate this process, the time and complexity associated with VM creation and associated storage management will be greatly reduced.

  • Automated failover capabilities. Having a storage system go offline has serious repercussions even without factoring VMware into the equation. But as organizations virtualize their servers and place them on an external storage system, the ramifications of that storage system going offline at any time becomes unacceptable. So to mitigate those risks, the FileStore N8300 v5.7 gives organizations the option to automate failover so organizations can recover at other sites.

Using these new automated failover functions in FileStore N8300 v5.7, organizations may configure replication jobs, users or both to failover to a secondary location and resume operations with minimal or no application interruption. Here again organizations will first need to set policies so the replication of data and automated failover occurs in accordance with their business needs. But the tools that they need to setup and manage this process are now natively included with FileStore N8300 v5.7.

s of all sizes have in recent years implemented VMware but are dealing with the reality that the storage arrays on which they place their VMs are not optimized either for a cost or performance perspective for the demands that VMware vSphere puts on them. The Symantec FileStore N8300 v5.7 provides this new breed of storage system that organizations need. By adding in automated failover, primary data deduplication and vCenter integration, organizations get the application availability they need, the cost savings they seek and the ease of management that they desire in the virtualized environments that they are creating.


Click Here to Signup for the DCIG Newsletter!


DCIG Newsletter Signup

Thank you for your interest in DCIG research and analysis.

Please sign up for the free DCIG Newsletter to have new analysis delivered to your inbox each week.