Cloud management tools are beginning to give IT what it really wants -- flexibility – regardless of the environment.
This is according to admin who wanted to move VMware vSphere workloads into a public cloud, instead of being locked into a vendor via VMware's vCloud Connector or another VMware partner's cloud.
Multi-cloud deployments are becoming common in the enterprise, as IT admin mix and match provider offerings to best fit their needs.
"Multi-cloud is most of the solutions I work on these days and vSphere is there half the time," said David Linthicum, senior vice president at Cloud Technology Partners in Boston. "The fact is most enterprises moving to cloud will do so using multi-cloud architecture and benefit a great deal from multi-cloud management technology."
The ability to move workloads easily between private data centers is/was the missing step for cloud adoption at the enterprise level.
As multi-cloud deployments grow, so does the need to manage various deployments.
'Cross-cloud' seems to be on the fast track to becoming a necessary feature in any cloud management tool, according to Dan Sullivan, principal of DS Applied Technologies in Portland, Oregon.
"The ability to move jobs across clouds is becoming a commodity feature in cloud management platforms," he said.
RightScale vSphere appliance
While some companies struggle to modernize their IT departments, others recognize the need to move existing workloads into the cloud.
"One of the things that stood out [was] that traditional IT ops wasn't working for us, which is why we needed to switch to something more agile, hence RightScale," said Alain Gaeremynck, senior enterprise architect at the Yellow Pages Group (YPG) in Montreal, Canada.
RightScale Inc. recently released a self-service appliance that allows customers to move vSphere workloads across different vSphere infrastructures as well as across all major cloud platforms, such as Amazon Web Services, Windows Azure or Google.
The stateless appliance sits behind a customer's firewall and between vSphere in the data center and RightScale's Cloud Portfolio management.
"The ability to move workloads easily between private data centers was the missing step for cloud adoption at the enterprise level," said Jonathan Frappier, a blogger at virtxpert.com. "It's very intriguing and something I've been hoping would come out for a while."
The tool is designed for a company like YPG to take its time moving workloads from its on-premises vSphere to a hybrid cloud environment. YPG hasn't installed the vSphere tool yet, but plans to fully integrate it within a couple months.
YPG didn't have support for vSphere workloads with its other providers; the RightScale appliance lets it implement even more workloads from its data center to the cloud.
The company has upward of 200 virtual machines managed in one data center, which must integrate RightScale's appliance into its workflow -- a process that will take some adjusting, Gaeremynck said.
"It'll be a bit of a learning curve for IT ops," Gaeremynck said. "It will take a bit of time to integrate with all the subnets and all the various zones to define the policies. But it's more of a process thing than a technical thing."
RightScale's Cloud Management tool starts at $9,000 per year, while the vSphere appliance is an additional $12,000 per year. Other similar tools include Dell Enstratius, which offers vSphere support, as well as products from Racemi Inc., RiverMeadow Software Inc., ServiceMesh and HotLink Corp.
'Cross-cloud' workload hiccups
Even with these cloud management tools, the process of moving workloads from vSphere to the cloud can prove difficult.
"I would suggest that moving compute jobs is much easier than moving large volumes of data," Sullivan said. "Database applications, for example, can't just be moved by moving compute jobs. The same goes for big data and data warehouse processes, like ETL or report generation."
"Data is like an anchor that can keep a company tied to one provider's cloud," he said.
The question is whether the benefits of moving compute jobs outweigh the cost of moving data that needs to go with it.
The limitations also vary on the size of the workloads, Linthicum said.
"Moving them constantly will cause too much overhead on the infrastructure," he said. "You can't just move things around willy-nilly; you have to consider what those workloads are doing."