Determining protection tiers in virtual environments is tricky considering the wide range of protocols and storage types and the sheer number of virtual machines (VMs) that are spun up on a regular basis. That’s why IT pros who are well down the path of virtualization are starting to change the way they look at data protection and recovery methodologies for VMs.
It’s clear that VMs are becoming increasingly diverse and are increasing in size. They also may have special resource requirements around high I/O and rates of change. You wouldn’t have seen this class of machine being virtualized just a few short years ago.
The workloads on these VMs need to be isolated – they can’t just be placed into large pools of shared VM resources. They are being treated like physical machines running on a hypervisor. These VMs are also more critical and automation is needed to tie protection policies based on SLA classes.
The bottom line is that one size doesn’t fit all and it’s not just about VADP anymore. We need to consider multiple approaches to address these needs.
At VMworld 2013, Commvault’s Don Foster and Jonathan Howard walked a sell-out crowd through some best practices for what to consider when deciding how to leverage tech to change business processes in large-scale virtual environments.
For those of you who missed it, here a few of those best practices from Don and Jonathan:
- Don’t treat VMs like the hosts of yesteryear. Start with the application and apply the right data protection methodology, based on SLAs.
- Snapshots are great for near instant recovery of critical applications. However, there are other options for apps of lesser importance. Don’t use tier 1 protection for lower-tier apps. Sometimes VADP is good enough for tier 2 and tier 3 applications.
- The reason we back up data is so we can recover it. Change the way you look at recovery. Hardware-based snapshots minimize the time it takes to quiesce and snap VMs.
- Whether it’s VADP, replication (for business continuity) or driving hardware snapshots, the key is to have options. Commvault’s IntelliSnap technology provides all of the above.
VMworld 2013 day three is behind us, but we’re not done yet. Swing by booth #1605 to see our demos and be sure to check out our remaining breakout sessions:
Breakout: Reigning in VM Sprawl – Automating the Reclamation of Unused Resources
Time: Wednesday, Aug. 28, 1 – 2 p.m.; Moscone West, Room 3012
Presenters: Jonathan Howard, John Fox
Session ID: VCM5827
Why go: It’s so easy today to provision new virtual machines, that many organizations are experiencing VM sprawl: the proliferation of virtual machines, many of which are only needed for a certain time, and are then orphaned. We’ve been talking about VM sprawl for years, and it never seems to get resolved – until now. We’ll examine ways to mitigate sprawl and learn about brand new tools to help VMware admins and Dev/Ops teams automate resource utilization of the data management layer, like VM Power Management (to reclaim unused memory and compute resources), Storage Management (to reclaim tier 1 disk space), and VM Archiving (to reclaim unused compute and storage resources).
Breakout: Drying Out After Hurricane Sandy: Leveraging vCloud Director and Other Cloud Enablement Tools for Simpler, Faster DR Operations
Time: Thursday, Aug. 29, 10:30 a.m. – 11:30 a.m.
Presenters: David Ngo, Phil Curran
Session ID: BCO5829
Why go: A big lesson learned from Hurricane Sandy is that DR readiness is not a one-time thing, but a daily struggle to ensure both recovery readiness and compliance. The proliferation of virtual infrastructures gives you many more tools to be DR ready and this session will discuss the emerging technologies that are the foundation of a successful business continuity plan. Making the short list are: WAN-optimized copy technologies, like deduplication-aware replication, VM encapsulation that leverages P2V and V2V conversion, and orchestration and automation capabilities that can understand the dynamics of both virtual and cloud-based models for DR.