The dws Data Plugin
Collects DataWarp stats from compute nodes
The dws plugin collects the following DataWarp utilization statistics from compute nodes, if presesnt. This usage data is available within the scope of an application, not the scope of a job. The data is written in JSON dictionary format. Additional DataWarp usage data is available through the dws_server and dws_job_server plugins.
bytes_read- Number of bytes read
bytes_written- Number of bytes written
files_created- Number of files created
inodes_created- Number of inodes created, including files, directories and links
max_read_offset- Maximum byte offset read
max_write_offset- Maximum byte offset written
RUR dws output
This example shows dws data as written to /var/opt/cray/log/partition-current/messages-date on the SMW:2017-02-03T13:27:32.662733-05:00 c0-1c0s8n0 RUR 16521
p0-20161006t064726 [RUR@34] uid: 12345, apid: 1140449, jobid:
1127.sdb, cmdname: /bin/hostname, plugin:dws {"token": "1127.sdb",
"mountpoint": "/var/opt/cray/dws/mounts/batch/1127.sdb/ss",
"inodes_created": 407, "files_created": 405, "bytes_read":
11207405004, "bytes_written": 6712208222, "max_read_offset":
4096024126, "max_write_offset": 21772241122}