Dinesh Singh

Why VVol for All-Flash systems?

Blog Post created by Dinesh Singh on Nov 19, 2015

Last week during a customer conversation on implementing VVol on VSP G1000, I was asked why anyone would need VVol management for all-flash arrays. Since all-flash arrays tend to provide only a single tier of physical storage, this limits the potential to differentiate resource groups and deliver VM specific QoS. I explained to him that a storage admin can differentiate storage containers (or data pools in Hitachi lingo) using many other capabilities of the underlying storage arrays. It could be any combination of capabilities including performance (IOPs, latency), availability (local or remote replication), snapshot, encryption, space-efficiency (deduplication, compression) etc. Even if we consider just the physical media, there will always be some storage classifications within an enterprise. It could be a mix of all-flash and hybrid storage or just the all-flash storage environment, which uses commercial SSD and Hitachi FMD, so VVol mechanism can help create differentiated storage pool across multiple systems.

 

Ability to deliver individual service levels per VMDK is only one aspect of the VVol framework. Another key benefit that can be delivered with VVols comes from Hitachi’s rich policy-based management (PBM) engine which ensures automation of VM provisioning per individual storage service requirement and also eliminates any possibility of human-induced error in assigning storage resources. This includes preventing errors like a storage type mismatch or resource overprovisioning. PBM from Hitachi  is of increased importance in certain use cases such as private and public cloud where storage admins can define storage containers for each tenant based on their specific VM requirement and show chargeback using Hitachi Command Suite’s unique cost capability (Shown in image 1). Hitachi Command Suite can also help in preventing non-critical VMs from monopolizing tier 1 storage services by exposing cost capability to vSphere admin and enforcing compliance as per end-user’s willingness to pay .

 

Pic 1.png

 

 

Coming back to my discussion about the value of VVol on an all-flash array, the customer then shared that his curiosity was largely because no AFA vendor currently supports VVol.

 

Introducing the VSP F series – the first AFA to support VVols

Last week we introduced the new VSP F series all flash arrays, which delivers traditional Hitachi reliability with a 100% data availability guarantee and are built on new FMD DC2 flash modules to deliver best  in class performance with no-penalty, inline compression. It is a big milestone in Hitachi’s endeavor to help customers modernize the IT core with software-defined solutions and strategic flash deployments.

Pic 2.png

 

 

True to Hitachi tradition of delivering nothing but the best to our customers, we are providing VVol integration on our newly launched all-flash VSP F series systems (VSP Fx00 for short hand) from day one. It took strategic partnership with VMware and years of engineering efforts to be the first all-flash storage partner to provide VVol integration.

 

VSP Fx00 is certified for VMware server virtualization / desktop virtualization use cases and supported with complete vRealize integrations (vRO, vROps, vRLI etc), providing simplified management for VMware environments including vRealize private and public clouds.

 

Pic 3.png

 

 

Unified VVol Management Across AFAs, Hybrid Arrays, SAN and IP protocols

 

To further simplify customer deployments and management of different storage systems, we are also announcing the general availability of a unified VASA provider 3.2 package that allows management of both block and file storage systems. Customers can now manage VSP G, F and Hitachi NAS systems by deploying just one VP package. We also introduced new release of VVol management platform Hitachi Command Suite (HCS) 8.2.1, which allows customers to define storage capabilities, create storage containers and manage other storage operations for block and file systems from unified interface. Customers can now expose similar set of capabilities for NFS, SAN and iSCSI protocol supporting storage arrays.

 

Stay tuned for more exciting solutions from Hitachi ! For more information on Hitachi’s solutions with VMware, please visit  HDS.com/go/vmware

Outcomes