Why scientists should care about storage
TRIUMF Lab Tier-1 Tape Library As the Large Hadron Collider restarts at Cern, data storage has become as important to
scientists as compute power. But, as Tom Wilkie reports, the innovative technologies being developed have much wider applications
C
ern, the European Laboratory for Particle Physics just outside Geneva, was the birthplace of the World Wide Web. Now, as Cern’s Large
Hadron Collider (LHC) starts a new round of experiments aſter the discovery of the Higgs boson in 2012, the laboratory is the crucible of yet more innovation that could change the face of computing in both science and the wider commercial world. Te LHC generated more than 100 petabytes
of data in its earlier runs, and will continue generating data at the rate of two to three petabytes each month. Tis huge quantity of data has to be distributed to scientists around the world so they can use it to discover new physics. Te way in which Cern has structured
22 SCIENTIFIC COMPUTING WORLD
its data distribution, and the techniques it and its associated laboratories are using to store that tidal wave of data, will have ramifications for the growth and development of big data in science and in commerce. According to Laura Shepard, director of
HPC markets, for storage company DDN: ‘Everyone knows that data is becoming more distributed and big data is key to large-scale endeavours not just in the scientific but also in the commercial community. It is about where the data is generated and where it is processed. Te Cern project is an amazing example of this, and it is only going to get more and more relevant as these large-scale scientific and commercial projects come on line. ’ In March, Seagate, the world’s leading hard-
disk drive company, announced that it had concluded a three-year partnership with Cern Open Lab to collaborate on the development of the Seagate Kinetic Open Storage platform, in the context of the LHC programme. Just as the scientists are having to learn about how best to store data in order that they can easily retrieve it for analysis, so the data storage companies are looking at the LHC project and cooperating with the laboratories in the distributed data network – DDN is working with the Canadian national laboratory, Triumf, in Vancouver, for example – to help them develop the next generation of storage technologies that will help not just science but more widely in the commercial world of big data as well. It wasn’t always like this. ‘Everyone talks
@scwmagazine l
www.scientific-computing.com
TRIUMF lab
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40