Cloud takes backup role as cities remake data centers
- By John Moore
- Jul 21, 2014
Third in a series.
With pressure on government to streamline IT operations to control costs and spur performance, a growing number of city IT managers are pursuing the latest hardware architectures and tools to renovate their legacy data centers.
But while the emphasis on physical devices may seem like a hardware renaissance, the cloud is still very much a part of data center overhauls. In such cases, agencies are using cloud in a backup role, a resource that soaks up spikes in data center workload or as a disaster recovery utility.
The city of Asheville, N.C. , for example, found that traditional disaster recovery involved a high, fixed capital expense. As a consequence, city’s budget only supported disaster recovery plans for applications it absolutely needed to protect – enterprise resource planning (ERP), for example, said Jonathan Feldman, Asheville’s CIO.
Plus, the protection for the application, exists only two blocks away from the city’s main data center. “I have been trying to jockey the funds to get our disaster recovery center farther away,” said Feldman, who joined the city nine years ago.
But now the city is moving its disaster recovery operations to the cloud, which offers a pay-as-you-go approach and usage-based pricing. The cloud lets Asheville enter disaster recovery mode for a couple of days and a few hundred dollars, versus the ongoing hardware costs of operating a physical disaster recovery facility. “That is something we can afford,” Feldman said.
Ashville leverages Amazon Web Services for disaster recovery. The city also uses software from CloudVelox Inc., a Santa Clara, Calif.-based company, to replicate data to the AWS cloud. CloudVelox provides an automated cloud migration disaster recovery platform.
Asheville turned to the cloud after a plan to build out a $200,000 disaster recovery center as part of a fire station construction project failed to materialize. Through the CloudVelox/AWS setup, the city was able to extend disaster recovery to applications that previously had to do without it.
Recent examples include an asset management application for tracking work orders and a point of sale (POS) application supporting the city’s 7,500-seat US Cellular Center. As for the latter, a POS outage could cause the center to lose $20,000 to $30,000 for a given event, Feldman noted.
Asheville plans to protect other applications in AWS. And while the city is looking at its ERP system, it must first conduct a due diligence process to vet security, he said.
Vijay Sammeta, CIO for the City of San Jose, Calif., also has plans for the cloud as a backup mechanism. He said the city will focus over the next 12 to 18 months on moving virtual machines to the cloud to manage a range of issues including backups and disaster recovery.
“When you think about all the components of a highly available service delivery stack: network, servers, database and the applications, it starts [to] make a lot of sense to simply let someone else worry about that and just build redundancy to the Internet,” Sammeta said.
In Newington, Conn., the town is testing a SimpliVity Inc. feature that would also extend the town’s disaster recovery and business continuity capabilities to AWS, according to Paul Boutot, Newington’s chief information officer.
The cloud can also provide a different flavor of backup by supplementing over-taxed data center resources. Oakland County, Mich. plans to provide some room for growth as it revamps its data center, but won’t attempt to provide enough capacity to meet its ultimate peak needs, CIO Phil Bertolini explained.
For peak volumes, the county plans to rely on cloud computing. “We will have some cloud components that we can ramp up or down as needed,” he said.
Previous: County mixes data center options with reference architecture