Fresh from its recent sinister triumph, the Wellcome Trust says that its research into genetic diseases is generating such huge data volumes that it has had to buy an extra 42TB of SATA disk arrays, 30TB of which are already full.
Scientific analysis of genetic material generates output files of around 50MB, the trust said, and these are kept on-line so they can be accessed by its partners in other research groups around the world.
It added that the data yield from its Wellcome Trust Centre for Human Genetics (WTCHG) has risen from 20GB a day a couple of years ago to between 200GB and 300GB a day now. At that rate, the remaining 12TB will fill in two to three months, and more arrays will be needed.
Based at Oxford University, WTCHG is part of a world-wide collaborative programme which is researching the genetic causes of diabetes, obesity and other common ailments. Its own compute resources include a 120-node Linux cluster and 25 core servers, plus a Fibre Channel SAN.
It also has a server that now hosts four 21TB Nexsan SATABeast arrays, mirrored for a total of 42TB, said Dr Tim Bardsley, WTCHG's IT manager. It manages its storage using DataCore's SANmelody software, which allows users to access data via iSCSI and Fibre Channel.
Along with low cost and power consumption, a prime factor in choosing the Nexsan storage was reliability, Bardsley said. "We have research programmes that have been running now for three years or more," he added, "and you cannot put a price on how valuable that data is."®