Convey Computer Increases Graph500 Performance, Accelerates Big Data Analytics
Convey Computer Corporation has announced two new entries on the Graph500 List. The entries increase performance three to four times over Convey’s prior results. Convey credits the dramatic speed up to the company’s recently developed breadth-first search (BFS) implementation and to the increased processing power of the newly introduced HC-2 Series.
Convey hybrid-core systems use personalities — customized instruction set architectures that increase performance of specific portions of an application, in this case the BFS algorithm. Using the new personality, the single-node Convey HC-1ex increased performance from 1.7 GTEPS to 5.9 GTEPS (billion edges per second), an increase of 3.5x on problem size 27. On the new HC-2ex, the performance was even more dramatic, clocking in at 7.8 GTEPS for an increase of 4.5x over the earlier HC-1ex. This kind of performance in a single-node system reinforces Convey as a performance/power leader in executing graph type applications.
Convey’s latest version of the BFS is a highly threaded algorithm that utilizes over three thousand independent threads of computation. This capability allows for massive parallelism that speeds all graph type applications. In addition, the unique hybrid nature of the architecture allows portions of the benchmark to simultaneously execute across the multiple compute resources of the system.
In April 2012, Convey announced a new family of hybrid-core systems — the HC-2 Series. Accelerating computing even more than previous Convey systems, the HC-2 provides users with greater performance, functionality, and energy efficiency. The Convey HC-2 systems increase application performance 2-3 times over previous generations of Convey servers and orders of magnitude over commodity servers.
“Graph applications are about finding and mapping links and relationships, and they are critical in understanding large amounts of unstructured data,” states Addison Snell of Intersect360 Research, an analyst firm highlighting the application of high-performance technologies to big data analytics. “Strong performance on the Graph500 benchmark indicates that the system could efficiently be applied to a wide range of complex analytics problems.”
Convey’s hybrid-core computing architecture tightly integrates advanced computer architecture and compiler technology with commercial, off-the-shelf hardware — namely Intel Xeon processors and Xilinx Field Programmable Gate Arrays (FPGAs). The systems help customers dramatically increase performance over industry standard servers while reducing energy costs associated with high-performance computing.
