Comprehensive data protection for all workloads
Post Reply
btkrausen
Influencer
Posts: 20
Liked: 2 times
Joined: Dec 29, 2009 7:05 pm
Full Name: Bryan Krausen
Contact:

Proxy Configuration for Multiple Clusters

Post by btkrausen » Mar 05, 2013 3:13 pm

Hi guys,

We have three clusters in the same datacenter which utilize different storage groups on the SAN. For proxy server configuration, would you recommend three different proxies connected into the three different storage groups, or one bigger proxy with three different HBAs, each of which is a member of three storage groups? The servers don't host any sensitive data. One cluster is DMZ, one cluster is for QA/Dev, and the bigger one is corporate production. The main Veeam server lives in another datacenter.

Currently our proxy in this datacenter is SAN connected and a member of the bigger clusters storage group, while utilizing network mode for the DMZ cluster but with it's massive growth, we obviously want to change that.

The current proxy has a IBM DS attached for it's storage, however, if we go with multiple proxies, we would have to purchase additional storage appliances (one for each proxy).

Thoughts?

Vitaliy S.
Product Manager
Posts: 22984
Liked: 1556 times
Joined: Mar 30, 2009 9:13 am
Full Name: Vitaliy Safarov
Contact:

Re: Proxy Configuration for Multiple Clusters

Post by Vitaliy S. » Mar 05, 2013 10:10 pm

Hi Bryan,

If I were you I would most likely go with three separate proxy servers, because of the redundancy and load distribution reasons. Keep in mind that each job will consume CPU/Memory resources of the proxy server, so your single server should be "beefy" enough to allow running three jobs at the same time without any job performance degradation.

Thanks!

dellock6
Veeam Software
Posts: 5734
Liked: 1625 times
Joined: Jul 26, 2009 3:39 pm
Full Name: Luca Dell'Oca
Location: Varese, Italy
Contact:

Re: Proxy Configuration for Multiple Clusters

Post by dellock6 » Mar 06, 2013 10:10 pm

Another note for storage appliances: while creating "Veeam Pods" each with its own local storage, this in some way removes any possibility to distribute jobs among the several proxies you have, since each job will be linked to a specific proxy. I did few weeks ago a similar design for a customer, and we took a dedicated storage (netapp 2240 with 50tb) but we direct attached it to only 1 proxy, who then also gained the repository role. All other proxies are going to save to this repository, so jobs can be run by every proxy we have.
If the proxy holding the repository role dies, a quick mount on the netapp on another proxy, change each job to map the existing backup files and use the new repository as target.

Luca.
Luca Dell'Oca
Principal EMEA Cloud Architect @ Veeam Software

@dellock6
https://www.virtualtothecore.com/
vExpert 2011 -> 2019
Veeam VMCE #1

Post Reply

Who is online

Users browsing this forum: Bing [Bot], Google [Bot] and 41 guests