SYSTAP CEO Brad Bebee to Speak at SC15 on Overcoming Barriers of Graphs on GPUs
Austin, Texas, SC15 (PRWEB) November 16, 2015 -- Brad Bebee, CEO of SYSTAP, creator of Blazegraph, the industry’s first GPU-accelerated high performance database for large graphs, will speak at SC15, the premier event for the international supercomputing community. Bebee’s presentation, “Overcoming the Barriers of Graphs on GPUs: Delivering Graph Analytics 100X Faster and 40X Cheaper”, is scheduled for Wednesday, Nov. 18, at 1 p.m. in the NVIDIA GPU Technology Theater (Booth 1021). SC15 runs November 16-19 at the Austin Convention Center in Austin, Texas, where SYSTAP is being featured by NVIDIA as a leading provider of data analytics solutions using GPUs.
While graph problems may look similar to other big data challenges, graphs have very different computational workloads and scaling requirements. Techniques that work on a small scale will typically fail to deliver on larger graphs, and because the volume of data under analysis is constantly increasing, it is virtually impossible to predict how to efficiently store the information for the rapid computational scale that will be required for ongoing exploration and discovery of data.
“Virtually every graph database application is eventually hampered by query and analytic performance as the data scale increases,” said Bebee. “To achieve graph analytics at scale, you need the right combination of hardware and software. Multi-GPU acceleration is an exciting addition to our existing High Availability and Scale-Out architectures. I look forward to explaining how Blazegraph can help speed data analytics and overcome other adoption barriers for applications of all sizes.”
For organizations and analytic application providers focused on heavy data mining and predictive analytics involving very large datasets with exponentially growing sets of relationships, existing graph databases using key-value stores or CPU in-memory approaches simply cannot provide the scale for aggressive computational analysis. SYSTAP has developed its technology to exploit the advantages of GPU hardware, providing the ability to deliver analytics 100 times faster and 40 times cheaper than CPU-based solutions. During his presentation, Bebee will discuss how SYSTAP’s Blazegraph GPU delivers 200x to 300x query performance increases for graph query applications while requiring little or no code changes and offering support for existing APIs. He will present results showing how Blazegraph DASL (available later this year) delivered 1800x speed-ups over Spark and GraphX graph traversals.
Blazegraph solutions power graphs for Fortune 500 customers, governments and research organizations, from drug discovery to device management to national defense. EMC powers its Viper Storage Resource Manager (SRM) solution using Blazegraph, and Wikimedia selected Blazegraph over Titan, Neo4j, and many others to power the Wikidata Query Service.
About SYSTAP
SYSTAP, LLC, is a provider of highly scalable software for solving complex graph and machine learning algorithms. Founded in 2006, the company is the creator of Blazegraph, an ultra-high performance graph database supporting up to 50 billion edges on a single machine, and Mapgraph. Both solutions use the new disruptive technology of graphics processing units (GPUs) to accelerate data-parallel graph analytics. SYSTAP’s solutions enable extreme scaling that is thousands of times faster and 40 times more affordable than CPU-based solutions. Fortune 500 companies and government agencies – including AutoDesk, DARPA, EMC, Wikimedia Foundation and Yahoo7 – rely on Blazegraph for graphs at scale because, in graphs, size matters. For more information, visit http://www.blazegraph.com.
Laurie Gibson, Kickstart for Blazegraph by Systap, http://www.blazegraph.com, +1 650 346 8990, [email protected]
Share this article