Five years ago today, an initial 4.5 petabytes of high-performance storage
was brought online, signalling the ‘birth’ of JASMIN and revolutionising
access to environmental science data for NERC. JASMIN (link opens in a new window)
is half super-computer and half data-centre and provides a globally
unique platform at STFC, hosted for computing and storing UK and
European environmental science data.
A bustling hub of scientific collaboration, JASMIN has nearly 150 NERC-related science projects using its group workspaces (link opens in a new window) and the CEDA archive (link opens in a new window). These workspaces bring together researchers from across the domains of data-intensive environmental science.
Around 1000 virtual machines and 5000 cores of computing power
are woven together with a fast internal network; which efficiently moves
petabytes of data around, into, and out of JASMIN every month. Today,
JASMIN’s storage capacity has tripled to over 16 petabytes - as a
comparison, 1 petabyte of average length songs would equate to
approximately 2000 years of continuous play!
Already, JASMIN has established its own ‘data gravity’ - the
ability of data to attract additional applications and services.
Research projects benefit (and in many cases have only become feasible)
from locating their analysis activities alongside the CEDA archive and
other resources, so that they can share their work efficiently with
colleagues and other projects alike. Meanwhile the JASMIN cloud has
evolved to enable projects to showcase their work via outward-facing
services which they maintain themselves.
The UPSCALE (link opens in a new window)
project is a great example of how complex science questions were able
to be answered via the use of JASMIN; processing time for climate model
analysis was reduced from 3 months to merely 24 hours - the size and
scale of this project would not have been possible without JASMIN.
JASMIN now has over 1200 users from a variety of scientific
backgrounds and levels of expertise. Over 100 (mostly doctoral) students
register to use JASMIN every year; with around 2000 more registering
each year to exploit data via the CEDA download services. A variety of training (link opens in a new window) courses are provided by the CEDA and JASMIN teams, to help users manage, access and visualise data.
Over the last five years, JASMIN has enabled a vibrant and
dynamic UK research community to do new and exciting science.
Significant but exciting challenges lie ahead, as the NERC community’s
demand for JASMIN’s data storage and computing continues to grow.
Happy 5th Birthday JASMIN!
The CEDA and STFC Scientific Computing Department teams