SRM alernatives

Anyone have any alternatives instead of using SRM (site revovery manager) with vmware 4.1?
CR5150Asked:
Who is Participating?
 
coolsport00Connect With a Mentor Commented:
Sure - basically, my jobs, beit backup or replication are based upon our VM (server) SLAs. It's criticality level determines 1. if it gets backed up locally and 2. if it gets replicated; then the criticality level also determines how often. So, I have my most critical VMs replicating daily, which are few VMs. Then, my next most critical that don't change much data-wise, get replicated twice/wk. And then my 3rd most critical that rarely, if ever, changes gets replicated once/month. For 'local' recovery, I solely do backups. We can afford a 'wee bit' of downtime. I can have a VM recovered in under 30mins from a b/u because of Veeam's de-dup technology. So, for my backups, I use the same logic (daily, bi-weekly, monthly), based upon 1. criticality level and 2. the potential for data change on the VM (server). Honestly, it took several meetings with my Director on how best to utilize our storage and yet have a fairly current restore of mission-critical VMs. Keep in mind as well, for our most critical of VMs (Exchange, SQL, etc.), after a VM is powered on that may be a day or 2 old, what I would do to get current is use our normal backup/restore software (Avamar) to restore data/DBs. That is because our DBs are stored on our VMs via RDM. If you have them on a standard virtual disk, you could use Veeam for everything.

Hope that helps.

Regards,
~coolsport00

p.s. BTW...to further answer your question about 'same time replication'...yes, you are right; you can't do them concurrently. So, you just have to time your replication job, add a little bit of time, then schedule your 2nd job to start, and so forth. I did run into that problem myself where a job of mind (beit back up or replication) would fail due to a job currently running.
0
 
coolsport00Commented:
I certainly do - Veeam Backup & Replication. I haven't used SRM, so I think it's a bit different in functionality, but I believe what you want is a DR solution, correct? What I do is use Veeam to replicate mission-critical VMs to a remote DR site. If I lose my main site DC, all I have to do is simply power on my replicated VMs and I'm up and going. I, of course, also use Veeam to locally backup my critical VMs in case they get corrupted for some reason. So, all you would need is a remote ESX/ESXi host, some kind of storage for datastore (I use a direct-attached MSA), and Veeam licensing.

Hope that helps.

Regards,
~coolsport00
0
 
Paul SolovyovskySenior IT AdvisorCommented:
Depending on the size of of your environment you can do a lot of SRM does manually. SRM works on top of SAN replication, if you have SAN replication already you can mount the LUNs manually at DR site and do the same type of testing.  If you don't have SAN replication you can use Veeam or Quest vReplicator that will do datastore to datastore replciation.
0
Improve Your Query Performance Tuning

In this FREE six-day email course, you'll learn from Janis Griffin, Database Performance Evangelist. She'll teach 12 steps that you can use to optimize your queries as much as possible and see measurable results in your work. Get started today!

 
CR5150Author Commented:
Veeam can do datastore to datastore replication? I thought it only did host to host.  So you are saying veeam an replicate at the SAN level?

Thanks
0
 
coolsport00Commented:
No...it replicates VMs *from* datastore to datastore...
0
 
CR5150Author Commented:
Also I do have San replication using HP cmc as was thinking of doing what SRM does manually, or scripting it to bring up the servers on the dr side (change ip addresses, etc)
0
 
CR5150Author Commented:
Right that's the problem I have with veeam is that it relies on snapshots.  I'm trying to get away from snapshots because I have so many going and when the shot is created and then deleted there is a slight, but noticeable pause that some of my users notice.
0
 
Paul SolovyovskySenior IT AdvisorCommented:
I have configured many SRM implementations and most are medium to large environments where SRM makes sense in that it automates the VM deployment at the DR site, provides an auditable test
0
 
CR5150Author Commented:
Coolsport.  I have veeam and am currently replicating locally. So you also replicate to your DR site.  How do you get around scheduling both, since you can't replicate a vm to two places at the same time.  I'd be interested to hear how you have it setup and what frequency you are replicating some or all of your vms to local storage and to your dr site (I'm assuming you must have multiple schedules that don't overlap).  This solution may actually work well for me because as you said, the replica is already in "standby" mode at the dr site so in the event of a disaster, all i would have to do is change the ip address of each vm (dr site is different network) and bring the vms up.
0
 
coolsport00Commented:
Ok...so my next question is are you doing backup and/or replications during business hours? Doing so after hours would obviously rectify any potential hinderance. Although, I will say this, my replication jobs can last several hours. Even though they are started overnite, they can continue on well into the day...

You are the 1st person in many posts I've assisted on with Veeam (or any VM b/u or replication product) that I've heard where creating/deleting a snap causes disruption to users. Hmm...interesting...

~coolsport00
0
 
coolsport00Connect With a Mentor Commented:
Another thing that played into our strategy is our traditional data backup/restore solution (Avamar). If a regular (data-based, not necessarily DB-based) VM was out of date file-wise, our most concern was first getting the VM up. Then, if I needed to restore files, I could do that with our b/u solution because we have a 'main site' Avamar node, and a 'DR site' Avamar node. So I can do restores at our DR site (the main and DR site replicates data between the 2). For our file storage, we have about 2TB of storage, so we can't use Veeam or our b/u solution for that. For that what we had to do is use a utililty like Replistor. I created a clone file server VM with a different name, then use replistor to continually replicate that data to the clone. The initial 'seeding' of data takes a couple to a few days, but after that, it's quick, and daily. So, we have a few technologies in place to do what we need. But, they weren't very expensive. We are a SMB (more towards the small size I think...about 60-80 servers (VMs), and about 250 employees).

~coolsport00
0
Question has a verified solution.

Are you are experiencing a similar issue? Get a personalized answer when you ask a related question.

Have a better answer? Share it in a comment.

All Courses

From novice to tech pro — start learning today.