金年会|金年会官方网站

Today, Sugon publishes a result of Sugon cluster with I620-G30 servers, which breaks two world records of performance and price/performance set by TPCx-BB,  in the field of Big Data technologies on TPC website. The performance that reaches a high big data query speed of completing 3383.95 BigBench queries per minute and the price/performance that reaches $307.86 for each BigBench query per minute are all ranked No.1 among all the published results in the world. The result reflects extremely robust data processing capacities and the most amazing cost efficient of Sugon servers. The following is a screenshot from TPC website(www.tpc.org):

TPCx-BB Express Benchmark BB (TPCx-BB) is a Big Data Benchmark which measures the performance of Hadoop-based Big Data systems. It measures the performance of both hardware and software components by executing 30 frequently performed analytical queries in the context of retailers with physical and online store presence.

“All the nodes in this cluster, including management nodes and worker nodes, are Sugon I620-G30 rack servers,  which are very suitable for many big data processing application scenarios as the upgrades of architecture, the optimization of platform and technical innovations on this product,”  the Sugon Vice President Qin Xiaoning said.

“Comparing with the previous world record, Sugon cluster with I620-G30 servers completes 3383.95 Big-bench Queries per minute, which is more than twice the rate of previous one and the cost for each query is only $307.86, which is almost half the pervious price/performance record,if regardless of the different scale factors between them. ” Qin said.

Qin said, the breakthroughs of this result reflects that Sugon servers make a milestone improvement on the maturity of productization and international standardization in computer industry.

For years, supported by the targeted programs  from the  national“863”project, Sugon is dedicated on researching and developing high performance computers andsuccessively developed the 100 TFlops-scale supercomputer and the 1000 TFlops-scale supercomputer.