[Orca-users] Measuring Tape Throughput

bias16 list2002 at bias.org
Fri Aug 9 12:22:21 PDT 2002


First of all, commenting out the following in orcallator.se seems to
work to add tape throughput data:

    // Block the listing of tape drives for now.
    if (RAW_disk[i].short_name[1] == 't' && RAW_disk[i].short_name[0]
== 's') {
      continue;
    }

Second, In addition to adding to ALL disk throughput graph, it also
create disk_runp graph which you can view by editing your
orcallator.cfg file to address the disk_runp_st# entries that are
added.

< data              
disk_runp_((?:c\d+t\d+d\d+)|(?:[ms]d\d+)|(?:st\d+))
< data                 
disk_runp_((?:c\d+t\d+d\d+)|(?:c\d+d\d+)|(?:[ms]d\d+)|(?:st\d+))
---
> data                  disk_runp_((?:c\d+t\d+d\d+)|(?:[ms]d\d+))
> data                 
disk_runp_((?:c\d+t\d+d\d+)|(?:c\d+d\d+)|(?:[ms]d\d+)


The goal of gathering this data was to get an idea of how much
throughput our tape units are getting.  Since 99% of the writes are
tape on said system, this works great to accomplish that.

I was wondering, however, if anyone has done any coding to get
throughput PER DISK graphs.  I imagine I would have to so some math on
the throughput graph and the percentage graph data to create a new set
of data.

- Liston



More information about the Orca-users mailing list