Think You’re Fully Virtualized? Think Again


By Graham Smith, Primary Data Virtualization Products Director

GrahamSmith-326x245 Think You're Fully Virtualized? Think Again

VVols and Data Virtualization Fuel New Breakthroughs in Efficiency

A staple of modern datacenter practices, server virtualization has evolved and matured to ease application and VM management while greatly increasing efficiency. Meanwhile, technologies supporting storage for virtualized server shave lagged behind, holding ITback from achieving greater potential. VVols (Virtual Volumes), one of the most exciting features in the VMware vSphere 6 release, gives datacenters the virtualized storage support needed to take the next step.

Let’s take a look a closer look at how emerging data virtualization technologies make it easier to adopt VVols and to further enhance data management capabilities.

VVols Give Power and Efficiency Back to the Datacenter

With vSphere 5.5 and earlier, VM storage can only be provisioned at the LUN or Volume level. As a result, to meet the Service Level Agreements (SLAs) of widely differing VMs, enterprise IT has to manage multiple data stores in vCenter across multiple storage devices. To ensure SLAs and protect business continuity, it has become standard practice for IT to overprovision storage by about double the VM requirements. The resulting storage sprawl is strains enterprise budgets.

Virtual Volumes (VVOLs) solve these problems by enabling organizations to provision, monitor and assign VM Storage Police sat the virtual disk level. This granular control over VM storage aligns VMDK data needs with storage capabilities.

The Cause of Slow VVol Adoption

Given such significant benefits, it may be surprising that enterprises have been slow to upgrade to vSphere 6 and deploy VVols. A primary reason for this is that an upgrade to VVols requires the use of VVol-certified storage. In many cases, this means companies have to replace or upgrade existing storage, which can be expensive. A second reason for slow adoption is that upgrades and migrations are typically a complex and time-consuming processes, taking time and resources away from other business initiatives. Finally, moving to VVolcapable storage may require new training for storage admins. This is something many enterprise IT departments are working on, but willtake time.

Data Virtualization Eliminates VVol Adoption Hurdles

The most promising solution to slow VVol adoption are technologies that facilitate the upgrade by virtualizingdata. By abstractingdata from underlying storage hardware, these solutions make it possible to access data within a single VVol-certified dataspace. This eliminates the need to upgrade hardware in order to use VVols.

Another important way data virtualization accelerates adoption is that it serves as a Universal VMware VASA provider, which gives IT just one VVol data store to manage. This greatly reduces complexity, makes it easier to adapt to new VVols workflows, and simplifies system maintenance requirements.

Proactive Data Management to Improve Service Level Compliance

In addition to streamlining migration, data virtualization enables a host of capabilities that make the datacenter more agile, flexible and cost-efficient.

Guarantee VM Performance at the Lowest Cost

Through VASA, VM Storage Policies can be mapped to data virtualization policies. This allows intelligent software to automatically redistribute workloads to maximize resiliency, cost efficiency, data protection, security and performance, as workloads change. By dynamically moving Virtual Disks to the lowest cost storage that meets VMSLAs (Service Level Agreements), enterprises can guarantee VM performance and regulatory compliance without the excessive over provisioning required in traditional environments.

Simplify Infrastructure by Aligning Resources with Demands

The ability to move data within a global dataspace also enables enterprises to make the best use of their infrastructure investments in the following ways:

  • Data virtualization simplifies administration by making all devices manageable by policy through a single pane of glass, integrating with hypervisor managers, such as vCenter and Openstack through APIs.
  • Data resources become a global pool that administrators can scale linearly and incrementally instead of a proliferation of storage containers and over provisioned Tier 1 and Tier 2 storage that must all be managed independently.
  • Data virtualization can automate archival, identifying cool or cold Virtual Disks to delete or move to lower cost storage. By freeing capacity on expensive storage tiers, this allows future storage purchases to be deferred. This can be done transparently across multiple hypervisor environments, such as ESXi, KVM, Hyper-V, and others, so companies can automate a single and consistent policy-based management scheme across all virtualized and physical environments.

VVols & Data Virtualization Enable the Software Defined Datacenter

IT leaders are eager to realize the benefits of VVols, but for many, adoption is in the distant future due to the time and budget to purchase hardware, and then plan and execute a migration. Data virtualization can eliminate many of these hurdles to create an adoption fast track and to give enterprise IT the tools they need to take the next step toward achieving maximum efficiency in virtualization through a fully Software Defined Datacenter (SDDC).