Hello!
If the VM have 10 VHD disks and i set the concurrent streams to 1, i can run the backup at 170MB/s. Only is possible to run at more than 170MB/s if i run two or more disks at same time.
I think that if i can run a backup up to 400MB/s with 03 disks, why i can't achieve this same performance with only one disk ?
Is there a limit on the Veeam or Windows side ? Or a metric to apply in this cases ? We are testing a scenario where the VMs have dozen of disks and in some cases the disks need to be copied in a dedicated JOB.
-
- Service Provider
- Posts: 239
- Liked: 18 times
- Joined: Mar 23, 2016 5:57 pm
- Full Name: Diogo Campregher
- Contact:
-
- Product Manager
- Posts: 8191
- Liked: 1322 times
- Joined: Feb 08, 2013 3:08 pm
- Full Name: Mike Resseler
- Location: Belgium
- Contact:
Re: Maximum performance on a single disk
Hi Diogo,
First: Are you also running those 03 disks with 1 concurrent task?
First: Are you also running those 03 disks with 1 concurrent task?
-
- Service Provider
- Posts: 239
- Liked: 18 times
- Joined: Mar 23, 2016 5:57 pm
- Full Name: Diogo Campregher
- Contact:
Re: Maximum performance on a single disk
These 03 disks with 03 concurrent tasks.
-
- Product Manager
- Posts: 8191
- Liked: 1322 times
- Joined: Feb 08, 2013 3:08 pm
- Full Name: Mike Resseler
- Location: Belgium
- Contact:
Re: Maximum performance on a single disk
Diogo,
When you are doing these 3 disks with 3 tasks, if parallel processing is enabled (and I assume so) each task will work with 1 disk to improve the total backup speed and have the time for a snapshot to be open less long. That means that with that one task you are maxing something out (most likely I think the stream of data that lands on the repository, check at the bottleneck for the job) while with 3 tasks you will have items in parallel and as long as the resources can handle it you get better results.
Makes sense?
When you are doing these 3 disks with 3 tasks, if parallel processing is enabled (and I assume so) each task will work with 1 disk to improve the total backup speed and have the time for a snapshot to be open less long. That means that with that one task you are maxing something out (most likely I think the stream of data that lands on the repository, check at the bottleneck for the job) while with 3 tasks you will have items in parallel and as long as the resources can handle it you get better results.
Makes sense?
Who is online
Users browsing this forum: No registered users and 8 guests