Shell Oil Explores Hybrid Cloud IT Infrastructure to Lower Power Costs for Big Data

When Shell Oil’s power requirements grew 3.6 MW in a nine-month period at its three global data centers and 400 regional sites, the company’s IT subsidiary started investigating cloud-based capacity to lower energy costs, Computer World reported.

Part of the challenge for Shell Oil IT has been the upcoming deployment of a new seismic sensor. The device created with Hewlett-Packard explores old wells and areas for oil, compiling complex geological data sets into huge files. During device testing, a single file was 1 petabyte (1 million GB), Computer World said.

Shell IT has ten years to create a virtual infrastructure that can manage the huge files – it has 46 petabytes so far and 10,000 well sites to track – and make it profitable for the company to deploy its new sensors. The company’s goal is to create a hybrid cloud model, where some applications run in a public cloud and others through an on-premises private cloud.

Security is a concern, and it took more than a year to get internal approval to move corporate data to the public cloud. Shell is using Amazon VPC and within that has created a private section of the Amazon Web Services cloud. It is testing out Hadoop for big data analytics.

So far, the company’s in-house server infrastructure is 60 percent virtual, with another 40 percent on physical servers for legacy applications that can’t be supported on a virtual infrastructure.

According to the 2012 Energy Efficient IT Report from CDW, some 54 percent of IT managers said they have or are developing programs to manage power demand in their data centers. About 62 percent of respondents said they agree that cloud computing is an energy-efficient approach to data center consolidation, up from 47 percent in 2010.

CDW describes cloud computing as a possible “game changer” for energy efficient IT.

And Environmental Leader columnist Clemens Pfeiffer, CTO of Power Assure, says one of 2012’s top data center trends will be for organizations to create their own private clouds, while making greater use of public clouds for other secure services as well as “spill-over” capacity during periods of peak demand.