Tuesday, January 25, 2011

Big Data; the New EMC Vernacular for Large Volume, File Based, Rich Content Data.

IDC in their latest version of Digital Universe (IDC Digital Universe May 2010) forecast that data created on an annual base will increase from 1.2 ZB (Zettabytes) in 2011 to 35ZB in 2020 with 84% of the data created in 2020 being unstructured (file/object). A real world illustration of Jevons Paradox at work (Jevons Paradox states that advances in technology that improves efficiency and lowers cost increases the demand for the resource). So the harder the technologist works to simplify and drive down the cost of storage the greater will be the storage consumer’s appetite. Just think of 500 billion hours of TV content being available on the internet by 2015. (Screen Future)

Look for EMC’s new vernacular for unstructured data to gain traction. At their announcement last week they introduced the concept of two “uber-buckets” for data classifications, Enterprise Data and “Big Data”. This is a convenient segmentation that helps EMC put logic behind the positioning of their Enterprise storage products, VNX and VMAX and their “Big Data” primary storage solutions, Isilon and Atmos.

Big Data is unstructured data, generated from such sources as virally growing mobile user applications, computer modeling, seismic, web applications, video and graphics, audio, PDF, satellite downloads, genetic sequencing, medical imaging etc. In short big data means very large data stores that are measured in multiple TB’s if not PB with incremental growth also measured in TB. With a projected CAGR (2009 to 2014) of 55% for file/object and 24% for block (IDC, May 2010, Enterprise Storage Systems) it is little wonder that EMC are lining up their solutions to position themselves as major participants in the “Big Data” space.

No comments: