Savvis offering competes directly with Amazon Web Services. Coming soon: daisy-chained data centers that allow systems in one locale to automatically fail over to another.
Service provider Savvis is now offering infrastructure as a service, SavvisDirect, that competes with Amazon Web Services. And while it hasn't done so yet, it may one day soon offer a competitive advantage: daisy-chained data centers that allow systems in one location to automatically fail over to another.
Savvis has been primarily a large enterprise supplier of cloud services, with specialized deployments built to order. It sold them based on long-term contracts, not by the hour as Amazon does. Now it's entering the on-demand, self-provisioning world of infrastructure as a service, with users able to select the amount of network, storage and CPU that they want with each server. It will sell such servers via hourly billing, the same as Amazon does.
The company launched SavvisDirect as a beta service on Oct. 10 and is still under review. It's slated to become a generally available service later this year.
SavvisDirect President Andrew Higginbotham said in an interview there were other differences between Amazon and his fledgling IaaS. "If the Gartner Magic Quadrant is real, then Amazon is in the upper right [as a leader] and everyone else is huddled in the bottom left," he observed in a tip of the hat to AWS.
Like Amazon, SavvisDirect will offer virtual servers based on the Xen hypervisor, with networking and storage, along with a separate long-term storage service.
But Savvis has tried to learn from the Amazon example and will offer a few differentiating characteristics. "We will offer instant chat for help, along with trouble tickets and support services" for the cost of the monthly service, services that can be obtained only through separate service contracts with Amazon, said Higginbotham in an interview.
It will also seek to leverage its position as a business unit of CenturyLink, the third largest telecommunications supplier in the U.S. Savvis was acquired by CenturyLink in July 2011 for $2.5 billion, which gives it access to CenturyLink's 207,000 miles of fiber optic networking. SavvisDirect will be hosted in data centers in Washington, D.C., and Santa Clara, Calif., with expansion into London, England, early in 2013. By mid-2013, it expects to offer the service from Singapore and Hong Kong.
As part of its highly redundant approach, Savvis will operate two data centers in each location "on opposite sides of the city" but linked by a network. When the service moves out of beta and becomes generally available late this year, Savvis plans to link its Washington, D.C., operations to its Santa Clara facilities by an asynchronous, low-latency network over CenturyLink lines. That would make it possible for a customer to be able to shift virtual servers as complex files from one coast to the other as a routine, on-demand cloud service, if desired.
As the massive Hurricane Sandy approached New York earlier this week, many cloud users might have wished for such a checkbox option. For a company with primary servers in New York City and backup servers in Virginia, it may have seemed possible that both might be knocked out at the same time, with no time left to make alternative arrangements.
Such a service could become routine under SavvisDirect, said Higganbotham.
"It will be interesting to have that option of deployment to either the East Coast or West Coast," he noted. "We don't want to see the kind of outages that [affect] the cloud providers today," he added.
Even with two data centers in each city, he said Savvis takes the additional step of building in multiple grids of servers with separate power supplies and telecom services. Amazon does the same thing in its data centers, calling the separate grids "availability zones."
Pricing was not immediately available but beta users will be able to see a price chart, company spokesmen said.
Savvis will also offer Microsoft Exchange, SharePoint and Lync as software as service.
Multicloud Infrastructure & Application ManagementEnterprise cloud adoption has evolved to the point where hybrid public/private cloud designs and use of multiple providers is common. Who among us has mastered provisioning resources in different clouds; allocating the right resources to each application; assigning applications to the "best" cloud provider based on performance or reliability requirements.
InformationWeek Must Reads Oct. 21, 2014InformationWeek's new Must Reads is a compendium of our best recent coverage of digital strategy. Learn why you should learn to embrace DevOps, how to avoid roadblocks for digital projects, what the five steps to API management are, and more.