Host-based backup of VMware vSphere VMs.
Post Reply
readie
Expert
Posts: 158
Liked: 30 times
Joined: Dec 05, 2010 9:29 am
Full Name: Bob Eadie
Contact:

Interesting speed difference

Post by readie »

I have two very similar backup jobs, each with 16 VMs, general purpose servers, using the same Veeam server and proxies and target. However, the stats are quite different . .
job 1 - Processed 1.6TB, Read 57GB, Transferred 22.7GB (2.5x), Duration 2:40:57, Processing rate 16MB/s, Bottleneck Target (22%, 4%, 51%, 54%)
job 2 - Processed 975GB, Read 170GB, Transferred 20.8GB (8.2x), Duration 1:08:35, Processing rate 54MB/s, Bottleneck Source (48%, 5%, 47%, 35%)

OK so the jobs are somewhat different, but job 2 reads over 3 times as much data, compresses at 8.2x rather than 2.5x, and Processing rate is over 3x as fast. Is the processing that much faster because the compression is much better for some reason?
What further diagnosis can I do to see why one is so much faster, or is it just down to the type of data on the VMs?
Bob
Bob Eadie
Computer Manager at Bedford School, UK (since 1999).
Veeam user since 2009.
foggy
Veeam Software
Posts: 21069
Liked: 2115 times
Joined: Jul 11, 2011 10:22 am
Full Name: Alexander Fogelson
Contact:

Re: Interesting speed difference

Post by foggy »

readie wrote:is it just down to the type of data on the VMs?
To a large extent. Various processing rate for different VMs is expected, think amount of changed blocks, their compressibility, inline deduplication, etc. There're also a number of other factors.

Btw, make sure CBT is enabled for both jobs, so you're comparing apples to apples.
Post Reply

Who is online

Users browsing this forum: No registered users and 88 guests