cc_staff
1,487
edits
National systems: Difference between revisions
Jump to navigation
Jump to search
m
→GP2 (SFU)
m (→Overview) |
m (→GP2 (SFU)) |
||
| Line 38: | Line 38: | ||
{| class="wikitable sortable" | {| class="wikitable sortable" | ||
|- | |- | ||
| '''Parallel filesystem:''' || Approximately 4PB usable capacity for temporary (<code>/scratch</code>) storage. | | '''Parallel filesystem:''' || | ||
Approximately 4PB usable capacity for temporary (<code>/scratch</code>) storage.<br /> | |||
Aggregate performance of approximately 40GB/s. Available to all nodes. | |||
|- | |- | ||
|'''External persistent storage:''' | |'''External persistent storage:''' | ||
|| | || | ||
10PB or more of storage for persistent (<code>/project</code> and other) storage. Available to compute nodes, but not designed for parallel workloads. | 10PB or more of storage for persistent (<code>/project</code> and other) storage.<br /> | ||
Available to compute nodes, but not designed for parallel workloads. | |||
|- | |- | ||
|'''High performance interconnect:''' | |'''High performance interconnect:''' | ||
|| | || | ||
Low-latency high-performance fabric connecting all nodes and temporary storage. The design of GP2 is to support multiple simultaneous parallel jobs of at least 1024 cores in a fully non-blocking manner. Jobs larger than 1024 cores would be less well-suited for the topology. | Low-latency high-performance fabric connecting all nodes and temporary storage. <br /> | ||
The design of GP2 is to support multiple simultaneous parallel jobs of at least 1024 cores in a fully non-blocking manner. Jobs larger than 1024 cores would be less well-suited for the topology. | |||
|} | |} | ||