Supercomputer firm Cray has unveiled an analytics system called Urika-GX that fuses the firm's supercomputing processing power with open source software to meet big data challenges.
The Cray Urika-GX is powered by Intel Xeon Broadwell cores, and comes with 22TB of memory, 35TB of local SSD storage and the Aries supercomputing interconnect.
Three enterprise-accessible configurations of 16, 32 and 48 nodes will be offered in a 42U 19in rack from Q3 2016, and larger configurations will be available in the second half of 2016. No prices have yet been given.
Dominik Ulmer, vice president of business operations for EMEA at Cray, told V3 that the platform is aimed at the growing number of firms handling huge data volumes but lacking the necessary tools to analyse the information at scale and speed.
Ulmer explained that the cloud or sprawling cluster servers are unsuited to this task. “The cloud may be good for proof-of-concept studies or if you want to do some big data analytics once in a rare moment, but if you need to make decisions often, and quickly, this offering is ideal,” he said.
“It’s important for data scientists to be able to test hypotheses on datasets and try different work flows and turn this around quickly so they can act on the data as necessary."
Ulmer added that the platform works with open standards by being compatible with Hadoop, Apache Spark, Apache Mesos and OpenStack.
“Because it runs in an open framework you can plug in your existing solutions, so you're not stuck in a closed box environment," he said.
One particular element of the Urika-GX system is the Cray Graph Engine which the company claimed can offer much faster returns on data queries.
The Broad Institute of MIT and Harvard, a non-profit research institute aimed at advancing the understanding and treatment of diseases, is currently testing the platform in beta.
“With the Cray Urika-GX, we had quality score recalibration results from our Genome Analysis Toolkit [GATK4] Apache Spark pipeline in nine minutes instead of 40 minutes,” said Adam Kiezun, GATK4 project lead at the Broad Institute.
“This highlights the potential to accelerate delivery of genomic insights to researchers who are making breakthroughs in the fight against disease.”
The new launch underlines Cray's efforts to expand from its heartland of giant supercomputer deployments usually ordered by governments or huge multi-nationals into other areas.
Cray said earlier this year that more enterprises are considering supercomputers as the amount of data they gather increases and the cost of deployment decreases.
Geoengineering on the sea floor near glaciers would form a new ice shelf to prevent melting
Alterations in capillary blood flow can be caused by body position change
Curiosity rover is in 'normal mode' but not transmitting scientific data back to base
NatWest outage comes a day after Barclays' IT systems shut out customers and staff