Evolution of Storage: VM-Aware Storage for Virtualization

Sachin Chheda is the Director of Product and Solution Marketing at Tintri; he has long been involved in Information Technology with positions at HP, NetApp and Nimble Storage--developing and taking to market products that power some of the largest enterprises. This is part two of a two part series about the mismatch between virtualization and storage.

sachin_chheda_tn

In part one, we discussed the mismatch between storage and virtualization in management, performance and scalability, data protection, and TCO and ROI.

Solving the mismatch between virtualization and storage can help IT with its efforts to move to 100 percent virtualization using software-defined infrastructure. The first requirement is for storage to understand and operate at VM-level. This means doing away with the traditional storage constructs of LUNs and volumes. This is one of the main concepts behind a VM-aware storage architecture, which represents an evolution of storage beyond SAN and NAS specifically for the virtualized datacenter.

Storage and Virtualization Need to be Intelligent

Operating with, and serving up, individual VMs is not enough. To make virtualization predictable for tier-one apps and end-user desktops and enable mixing and matching workloads, storage should be intelligent, to determine what data is active at the individual virtual machine (VM)-level and guarantee equitable performance or a quality of service (QoS) for all VMs.

Data reduction techniques to maximize the usage of expensive storage, such as flash, as complements to QoS, and hybrid flash plus disk storage setups are a must for virtualized environments. When properly paired with high capacity and dense disk storage, compression and deduplication of data on flash can cost-effectively deliver both performance and capacity beyond what flash-only storage can claim.

Monitoring Analytics Is Critical

Performance analytics are also a must for storage for virtual environments. Offering deep insight into the performance of individual VMs (not LUNs or volumes with multiple VMs) can simplify trouble shooting, help identify the impact of new workloads and detect any trends. Jeff Boles from the Taneja Group estimates that the impact of having an end-to-end view of VM performance can be measured in substantial time savings — on the order of days — every time administrators must troubleshoot performance issues in their virtualized environments.

Adopting a modular approach to scaling using VMs and virtual disks as the unit for deploying storage is now possible using virtualization functionality such as VMware Storage DRS to load balance across different storage systems. This greatly simplifies how administrators can scale their environment without the complexity of scale-out or scale-up storage solutions. Adding the ability to control and monitor individual storage systems from a centralized administrative interface can further reduce the overhead IT faces with storage.

Lastly, VM-aware storage must provide administrators the ability to protect individual VMs (and not LUNs/volumes) with efficient snapshots and WAN-efficient replication. Jason Buffington, a senior analyst with Enterprise Strategy Group has analyzed and extensively written about concept of snapshot for protecting data. In his blog from Oct 2012, “Snapshots vs. Backups — a great debate, no longer,” he offers the advice of first deciding how to recover and then picking the protection method. VM-aware storage must also offer the ability to elegantly and quickly recover individual VMs from backups in addition to enabling quick and efficient backups of individual VMs. Efficient VM-granular replication over the WAN is also a must to facilitate disaster recovery.

Using VM-aware storage with cost-effective performance and capacity, simplified management and scalability, deep insight, and efficient protection and disaster recovery can enable IT to achieve 100 percent virtualization with excellent ROI.

Industry Perspectives is a content channel at Data Center Knowledge highlighting thought leadership in the data center arena. See our guidelines and submission process for information on participating. View previously published Industry Perspectives in our Knowledge Library.