cc_staff
28
edits
Cedar: Difference between revisions
Jump to navigation
Jump to search
no edit summary
No edit summary |
No edit summary |
||
| Line 35: | Line 35: | ||
|'''High performance interconnect''' | |'''High performance interconnect''' | ||
|| | || | ||
Intel OmniPath (version 1) interconnect. A low-latency high-performance fabric connecting all nodes and temporary storage. <br /> | Intel OmniPath (version 1) interconnect (100Gbit/s bandwidth). A low-latency high-performance fabric connecting all nodes and temporary storage. <br /> | ||
The design of Cedar is to support multiple simultaneous parallel jobs of | The design of Cedar is to support multiple simultaneous parallel jobs of up to 1024 cores in a fully non-blocking manner. For larger jobs the interconnect has a 2:1 blocking factor, i.e., even for jobs running on several thousand cores the Cedar system provides a high-performance interconnect. | ||
|} | |} | ||
| Line 42: | Line 42: | ||
<!--T:17--> | <!--T:17--> | ||
Cedar will have a total of 27,696 CPU cores for computation, and 584 GPU devices. Total theoretical peak double precision performance is 936 | Cedar will have a total of 27,696 CPU cores for computation, and 584 GPU devices. Total theoretical peak double precision performance is 936 teraflops for CPUs, plus 2,744 for GPUs, yielding over 3.6 petaflops of theoretical peak double precision performance. 22 fully connected "islands" of 32 base or large nodes will each have 1024 cores in a fully non-blocking topology, with each island expected to yield over 30 teraflops of measured double precision performance (measured with high performance LINPACK). There is a 2:1 blocking factor between the 1024 core islands. | ||
<!--T:7--> | <!--T:7--> | ||