-
- Veeam Software
- Posts: 100
- Liked: 23 times
- Joined: Oct 03, 2017 12:41 pm
- Full Name: Mark Polin
- Contact:
How fast can VBR backup to, and recovery from, Pure FlashArray//C?
As the global solutions architect for Veeam's Pure Storage alliance, I have been asked this question many times. We now have performance results!
Active Full backup performance of 4GB/s
Full restore performance of 4.7GB/s
Check out the Pure Storage Reference Architecture white paper for more results and how the environment was configured:
https://www.purestorage.com/docs.html?i ... eprint.pdf
Also, this reference architecture aligns with the Veeam Security Blueprint for 1000 VMs and 100 physical servers. Check that out here:
https://community.veeam.com/alliance-ve ... 13&fid=150
Active Full backup performance of 4GB/s
Full restore performance of 4.7GB/s
Check out the Pure Storage Reference Architecture white paper for more results and how the environment was configured:
https://www.purestorage.com/docs.html?i ... eprint.pdf
Also, this reference architecture aligns with the Veeam Security Blueprint for 1000 VMs and 100 physical servers. Check that out here:
https://community.veeam.com/alliance-ve ... 13&fid=150
-
- Novice
- Posts: 3
- Liked: 1 time
- Joined: Oct 31, 2023 2:08 pm
- Full Name: Wayne Johnson
- Contact:
Re: How fast can VBR backup to, and recovery from, Pure FlashArray//C?
Any performance data on backup and recovery from a Flashblade/E, file or object based?
-
- Novice
- Posts: 7
- Liked: 2 times
- Joined: Sep 17, 2021 7:09 am
- Full Name: Simone
- Contact:
Re: How fast can VBR backup to, and recovery from, Pure FlashArray//C?
Can we have also the compression ratio?
-
- Veeam Software
- Posts: 100
- Liked: 23 times
- Joined: Oct 03, 2017 12:41 pm
- Full Name: Mark Polin
- Contact:
Re: How fast can VBR backup to, and recovery from, Pure FlashArray//C?
Hi @wejohnson
Unfortunately, FlashBlade as a source was not part of the scope of this test. I will pass along your request that FlashBlade be tested as a source with VBR.
Thank you, Mark
Unfortunately, FlashBlade as a source was not part of the scope of this test. I will pass along your request that FlashBlade be tested as a source with VBR.
Thank you, Mark
-
- Veeam Software
- Posts: 100
- Liked: 23 times
- Joined: Oct 03, 2017 12:41 pm
- Full Name: Mark Polin
- Contact:
Re: How fast can VBR backup to, and recovery from, Pure FlashArray//C?
Hi @s@simo77
Thank you for your question. I see that the white paper mentions VBR's data reduction of 1.2:1, but does not mention the data reduction rate of the FlashArray//C target. I have asked Pure for this information, and will provide an update as soon as I receive a reply.
Thank you for your question. I see that the white paper mentions VBR's data reduction of 1.2:1, but does not mention the data reduction rate of the FlashArray//C target. I have asked Pure for this information, and will provide an update as soon as I receive a reply.
-
- Veeam Software
- Posts: 100
- Liked: 23 times
- Joined: Oct 03, 2017 12:41 pm
- Full Name: Mark Polin
- Contact:
Re: How fast can VBR backup to, and recovery from, Pure FlashArray//C?
Hi @simo77
I checked with Pure on the data reduction they observed during the testing. Unfortunately, they are unable to identify specific backup job data reduction rates because many jobs and many backups were sent to the same FlashArray//C-based backup repository. The FlashArray//C used for backup storage in their testing reports a Data Reduction Rate (DRR) of 15:1. This is high because the array is deduplicating across all the different jobs.
One other point is that Pure used incompressible data (see page 12) for their testing.
Thankfully, Pure, through it’s Pure1 management software, collects information from participating customers on the Data Reduction Rate for various types of data stored on FlashArrays, including Veeam backup repository data. Pure routinely sees DRR with Veeam between 3:1 and 4:1.
I checked with Pure on the data reduction they observed during the testing. Unfortunately, they are unable to identify specific backup job data reduction rates because many jobs and many backups were sent to the same FlashArray//C-based backup repository. The FlashArray//C used for backup storage in their testing reports a Data Reduction Rate (DRR) of 15:1. This is high because the array is deduplicating across all the different jobs.
One other point is that Pure used incompressible data (see page 12) for their testing.
Thankfully, Pure, through it’s Pure1 management software, collects information from participating customers on the Data Reduction Rate for various types of data stored on FlashArrays, including Veeam backup repository data. Pure routinely sees DRR with Veeam between 3:1 and 4:1.
Who is online
Users browsing this forum: Baidu [Spider], Google [Bot] and 66 guests