Comprehensive data protection for all workloads
Post Reply
Yuki
Veeam ProPartner
Posts: 252
Liked: 26 times
Joined: Apr 05, 2011 11:44 pm
Contact:

Bottleneck calculations

Post by Yuki »

Just wanted to drop a note that i think bottleneck calculations/screen may not be serving up information that should be relied upon.

We have a few large servers being backed up and replicated with veeam on 1Gbit network in hot-add mode to other servers/NAS storage. Our Veeam server/proxies are 8 core, 8GB+ ram machines. In all local actions, Veeam was showing bottleneck as source or target (80-90%+) vs network/proxy (7-25%). We've just upgraded the servers and network to 10Gbit and our throughput more than doubled, completion times are much shorter and it still shows the bottleneck as source/target. Which makes more sense.

However, I think it is fair to assume that previously, the network was the bottleneck since it's upgrade increased the throughput drastically. Then we added a new test target for backups with faster disk system and are getting 400MB/s on the processing rate in the screenshot below with source being the bottleneck all the time - which is correct given it is a lonely RAID5 across 12 SATA drives on Perc6 controller which pales next to new $1,500 LSI controller and NL-SAS drives in the new storage box.

Image
Post Reply

Who is online

Users browsing this forum: BackupBytesTim, Bing [Bot], Semrush [Bot] and 272 guests