We've been running centos repo's to write to our nfs data stores, there are advantages this way, but with REFS becoming stable we've been doing some testing with that as well. REFS works great, but the speed to duplicate to tape is about 75% slower for some reason.
The jobs are the same reverse incremental.
Are there some undocumented tweaks that need to be done or is this a case of calling into support to get them to tweak it?
-
kspare
- Enthusiast
- Posts: 33
- Liked: 4 times
- Joined: Nov 29, 2018 1:18 am
- Full Name: Kevin Pare
- Contact:
-
Dima P.
- Product Manager
- Posts: 14415
- Liked: 1576 times
- Joined: Feb 04, 2013 2:07 pm
- Full Name: Dmitry Popov
- Location: Prague
- Contact:
Re: Duplicate to Tape Performance
Hello Kevin,
What are the bottleneck stats for this tape job? I would suspect target repository, as data needs to be re-hydrated prior writing it to tape. Thanks!
What are the bottleneck stats for this tape job? I would suspect target repository, as data needs to be re-hydrated prior writing it to tape. Thanks!
-
kspare
- Enthusiast
- Posts: 33
- Liked: 4 times
- Joined: Nov 29, 2018 1:18 am
- Full Name: Kevin Pare
- Contact:
Re: Duplicate to Tape Performance
The refs disk is on the same storage server as the nfs repo.
-
Dima P.
- Product Manager
- Posts: 14415
- Liked: 1576 times
- Joined: Feb 04, 2013 2:07 pm
- Full Name: Dmitry Popov
- Location: Prague
- Contact:
Re: Duplicate to Tape Performance
Hello Kevin,
Can you please share the bottleneck stats from the tape job (ideally for both Refs and NFS repository)? Thank you!
Can you please share the bottleneck stats from the tape job (ideally for both Refs and NFS repository)? Thank you!
Who is online
Users browsing this forum: No registered users and 20 guests