News Stay informed about the latest enterprise technology news and product updates.

Long distance vMotion in focus with VCE’s new VPLEX add-on

Long-distance vMotion is a hot topic in theory, but experts say use cases are still relatively limited, even among the large enterprises that can afford it.

A new add-on to the Vblock converged infrastructure will offer IT shops a way to live migrate virtual machines and data between geographically separate data centers.

The ability to move virtual machines (VMs) between different data centers isn’t new, but with EMC Corp.’s VPLEX added on to the VCE Company’s Vblock, users can buy a prepackaged bundle of products designed for long-distance vMotion.

While the package turned heads as it was demonstrated this week at EMC World in Las Vegas, experts say even the types of businesses that can afford long-distance vMotion don’t often put it to use.

“It’s really, really [on the] fringe,” according to Chris Wolf, research vice president for Gartner Inc., headquartered in Stamford, Conn.

One goal that’s been proposed for such a system is disaster avoidance – the ability to vMotion workloads away from a primary site in the case of a known disaster, such as a hurricane on its way. But “customers have figured out that that’s not even in the realm of possibility,” Wolf said. “The laws of physics just don’t let you move everything that quickly.”

An increasingly popular scenario for long-distance vMotion is to create an active-active data center model, where spare capacity in a secondary data center can be used to feed the primary location.

“For tactical things like load balancing and data center consolidation and migrations, it’s pretty useful,” Wolf said.

The idea of active-active data centers appeals to IT pros, and it may be technically straightforward to add on the VPLEX to a Vblock, according to Jon Shulda, systems administrator for Rotary International, an organization of service clubs based in Evanston, Ill.

The problem is that Rotary recently implemented its Vblock 300 model GX, based on EMC’s VNX array, and there would be a significant cost to upgrade.

“We could get to [active-active data centers], we would love to, but I don’t know if we’re going to have the ability given the cost to get to that level,” Shulda said. “The technology is definitely interesting … but I just don’t think it’s something we’re going to be able to justify in the near future.”

Instead, the organization may try something like VMware’s vCloud Director to spin up clones of production VMs and bring them online at a secondary site with minimal downtime.

VPLEX for Vblock details
Pricing for the VPLEX starts at $77,000 for the single-engine VPLEX Local system; a VPLEX Metro configuration, which is used to do vMotions over distance, requires at least two systems. Other costs for long-distance live migration can include a large network pipe between data centers – an IP network with a minimum bandwidth of 622 Mbps is required, according to VMware’s support documents. Cisco Systems Inc.’s Overlay Transport Virtualization must be used to stretch the Layer 2 domain between data centers so that vMotions can be performed.

Vblocks consist of VMware’s vSphere 5, Cisco’s UCS B-series blade servers, and EMC’s storage arrays. The VPLEX add-on to the Vblock 300 and 700 series systems would be located in the Vblock cabinet that is ordered through, built by and serviced by VCE.

The VPLEX engine is built on two Intel Xeon-based high availability directors, along with redundant power supplies, battery backup and remote support capabilities. VPLEX software provides a distributed federation capability, which allows for data consistency across geographic distance to keep up with live migrating VMs.

Beth Pariseau is a senior news writer for and Write to her at or follow @PariseauTT on Twitter.

Dig Deeper on Disaster recovery strategies, business continuity and virtualization

Start the conversation

Send me notifications when other members comment.

Please create a username to comment.