Faster Than CMU’s Supercomputer

GStream 2.0 can process data of 32 billion edges in 500 seconds and process data of up to 256 billion edges while the supercomputer-based GraphLab of Carnegie Mellon University requires 1,400 seconds to process graph data having up to 32 billion edges.
GStream 2.0 can process data of 32 billion edges in 500 seconds and process data of up to 256 billion edges while the supercomputer-based GraphLab of Carnegie Mellon University requires 1,400 seconds to process graph data having up to 32 billion edges.

 

The Daegu Gyeongbuk Institute of Science & Technology announced on July 7 that it developed GStream 2.0, which is characterized by enabling a single personal computer available at five million won or so to handle a huge amount of graph data even a supercomputer has difficulty in processing.

GStream 2.0 employs a PC in which two GPUs and two PCI-e SSDs are used for graph-type data groups. This PC processes 256 billion edges at a speed of 2 GTEPS and can be utilized in various fields such as the Internet of Things, brain science, artificial intelligence and social networking.

“The GraphLab of Carnegie Mellon University, which is the most advanced big data analyzer at this point in time, requires 1,400 seconds to process graph data having up to 32 billion edges with a supercomputer equipped with 480 CPU cores, a 2 TB memory and a 5 GB network,” the institute explained, adding, “However, GStream 2.0 can process data of 32 billion edges in 500 seconds and process data of up to 256 billion edges by storing large-scale graph data in the PCI-e SSDs, performing asynchronous streaming from the SSDs to the GPUs and using thousands of computing cores in the GPUs.”

Details of the research were announced at the 2016 ACM SIGMOD/PODS Conference that took place in San Francisco on June 28.

 

Copyright © BusinessKorea. Prohibited from unauthorized reproduction and redistribution