Raijin
The original Raijin system was installed in 2012 and entered production use in June 2013. System updates entered production in November 2016, and again in mid-2017. Raijin currently comprises:
- 84,656 cores (Intel Xeon Sandy Bridge 2.6 GHz, Broadwell 2.6 GHz) in 4416 compute nodes
- 120 NVIDIA Tesla K80 GPUs in 30 nodes and 8 NVIDIA Tesla P100 GPUs in 2 nodes
- 32 Intel Xeon Phi (64 core Knights Landing, 1.3 GHz) in 32 compute nodes
- 4 IBM POWER8 nodes (64 cores running at 4.02GHz)
- 300 Terabytes of main memory
- Hybrid FDR/EDR Mellanox Infiniband full fat tree interconnect (up to 100 Gb/sec)
- 8 Petabytes of high-performance operational storage capacity
Raijin is supported by 37PB of project storage space for big data.
The system is currently undergoing a process to refresh the hardware. More details will be available soon.
Raijin Storage
The following table briefly summarise the file systems on Raijin.
Name[1] |
Purpose |
Availability |
Quota[2] |
Timelimit[3] |
Backup |
---|---|---|---|---|---|
/home/unigrp/user |
Irreproducible data eg. source code |
Raijin only |
2GB /user |
none |
Yes |
/short/projectid |
Large data I/O; data maintained beyond one job |
Raijin only |
72GB /project |
none |
No |
/g/data/projectid[4] |
Processing of large data files |
global |
|
none |
No |
massdata[5] |
Archiving large data files |
external – access using the mdss command |
20GB |
none |
Yes: 2 copies in two different locations |
$PBS_JOBFS |
I/O intensive data over the job’s lifetime |
local to each individual Raijin node |
unlimited[6] |
duration of job |
No |
Last updated: 25 May 2019
Source: https://inter.fyi/TLLmz