asetek announced a new order from existing oem partner fujitsu for the national institute of advanced industrial science and technology (aist).
the order for asetek rackcdu d2c™ (direct-to-chip) liquid cooling will be used at the ai bridging cloud infrastructure (abci) cluster, set to become the fastest supercomputer system in japan.
the order has a value of usd 975,000 with delivery to be completed in q4 2017.
“the efficiency of asetek liquid cooling enables data centers to continually push the performance of today’s most powerful processors,” said john hamill, asetek chief operating officer.
“we are pleased to partner with fujitsu to, once again, cool the fastest supercomputer system in japan. the order is another example of our ability to increase end-user adoption with existing oem partners and to expand our leading position in the hpc segment.”
as the core of abci, the high-performance computational system will consist of 1,088 primergy cx2570 m4 servers, mounted in fujitsu’s primergy cx400 m4 multi-node chassis. each server will feature the latest components, including two intel® xeon® gold processor cpus (a total of 2,176 cpus) and four nvidia® tesla® v100 gpu computing cards (a total of 4,352 gpus).
the 2u size chassis primergy cx400 m4 can each mount two primergy cx2570 m4 server nodes with gpu computing cards, offering high installation density. in addition, by utilizing asetek hot water cooling for the servers, the system can also realize significant power savings.
if this system had competed in the latest top500 ranking of supercomputers from june 2017, it would have taken the top position in japan and third place globally. abci will be deployed to the new abci datacenter on the kashiwa ii campus of the university of tokyo and is scheduled to begin operations in fiscal 2018 (the fiscal year starts april 1).