ExaGrid
-
Anybody heard of it? Thoughts?
At a Veeam User Group and ExaGrid is sponsoring it. The salesman gives a nice presentation of course, but I want to hear from experiences people on this one.
-
Yes, they are a decently major storage player.
-
We used Exagrid at my former employer. Ours was paired with BackupExec for Exchange and file backups.
They were VERY expensive even back then (this was 10 years ago) but for the ease of restores and the early data dedupe features, could not be beat. Today would probably be looking at Veeam and some low cost NAS box. -
Their big selling point is extreme dedupe, which while good is very much an under the hood thing.
-
So I asked the guy "What would happen if a controller in one of their ExaGrid node had died?" He said that the only things that are user replaceable are the drives, the PSUs, and the chassis. If there were multiple ExaGrid nodes clustered together, and one dies, the data on that one would not be available to you until it was restored.
I wasn't too keen on that idea. I was thinking more like striping across different nodes and restoring the missing data from checksums of the data that it had, but he was saying that that data would just be missing.
Not sounding like a viable solution to me. I'll keep my Synology's.
-
So I asked the guy "What would happen if a controller in one of their ExaGrid node had died?" He said that the only things that are user replaceable are the drives, the PSUs, and the chassis. If there were multiple ExaGrid nodes clustered together, and one dies, the data on that one would not be available to you until it was restored.
I wasn't too keen on that idea. I was thinking more like striping across different nodes and restoring the missing data from checksums of the data that it had, but he was saying that that data would just be missing.
Not sounding like a viable solution to me. I'll keep my Synology's.
But the Synologys have the same limitation. Is losing a single node really a problem?
-
I wasn't too keen on that idea. I was thinking more like striping across different nodes and restoring the missing data from checksums of the data that it had, but he was saying that that data would just be missing.
That's how Veeam Scale Out works, too. It's not that this is a risky scheme, you are just thinking of it as being an HA cluster when it is only a scale out cluster. But since you didn't need HA before, why do you need it in just this case? And when do you ever need HA with a backup repo?
-
HA would require much higher cost that even big banks and such would not spend money on. HA primary storage, definitely. But secondary storage? These are enterprise devices with enterprise support.
-
@scottalanmiller said in ExaGrid:
So I asked the guy "What would happen if a controller in one of their ExaGrid node had died?" He said that the only things that are user replaceable are the drives, the PSUs, and the chassis. If there were multiple ExaGrid nodes clustered together, and one dies, the data on that one would not be available to you until it was restored.
I wasn't too keen on that idea. I was thinking more like striping across different nodes and restoring the missing data from checksums of the data that it had, but he was saying that that data would just be missing.
Not sounding like a viable solution to me. I'll keep my Synology's.
But the Synologys have the same limitation. Is losing a single node really a problem?
Probably not. I haven't lost a node of anything in my career, but it could just be dumb luck so far that that has happened, or not happened. I have always thought that redundancy in the datacenter is a good thing, depending on the needs of the company.
-
I have always thought that redundancy in the datacenter is a good thing, depending on the needs of the company.
Definitely no, it tends to be a bad thing. It's certainly never an "always do this" kind of thing. Redundancy isn't even what we care about, reliability is. Redundancy as its own reward is a marketing trick from unscrupulous sales people trying to oversell hardware.
-
@scottalanmiller said in ExaGrid:
I have always thought that redundancy in the datacenter is a good thing, depending on the needs of the company.
Definitely no, it tends to be a bad thing. It's certainly never an "always do this" kind of thing. Redundancy isn't even what we care about, reliability is. Redundancy as its own reward is a marketing trick from unscrupulous sales people trying to oversell hardware.
I have one house, but for redundancy I built a mirror one on the lot next door.
-
I haven't lost a node of anything in my career, but it could just be dumb luck so far that that has happened, or not happened.
Right, but your response to that doesn't make sense. You are currently using Synology which is more likely to lose a node than ExaGrid is (SMB vs. enterprise gear.) And the repair and recovery service on the ExaGrid is far faster and better (hours rather than days.) And with scale out, the risk of losing one node is lower than the risk of losing one node without scale out because in one case you lose only some of your data temporarily rather than losing all of it.
So your fear of lack of redundancy and your fear of node loss, which are probably not rational regardless of the solution, have led you to respond by REDUCING your reliability further by avoiding safer solutions. Because you applied your guideline of wanting redundancy asymmetrically. You turn down ExaGrid for its level of redundancy but accept Synology that has less.
-
And remember, backups are redundant already literally by definition. So having your backups offline simply means you have reduced redundancy. Having redundancy of the backups is generally overkill. If "redundancy is always good" we can just keep going with where to apply it.
-
@NerdyDad Check out what the difference is between a RAIN and a RAID.
-
We have one at work. I don't manage it but I've heard good things about it.
-
Anybody heard of it? Thoughts?
At a Veeam User Group and ExaGrid is sponsoring it. The salesman gives a nice presentation of course, but I want to hear from experiences people on this one.
This is what you really want/need:
Decent hardware platform + software you know how to use (Windows Server + Veeam) + support in truly fire-n-forget mode.
In the worst case you'll re-provision non-proprietary hardware after EOL.
-
So I asked the guy "What would happen if a controller in one of their ExaGrid node had died?" He said that the only things that are user replaceable are the drives, the PSUs, and the chassis. If there were multiple ExaGrid nodes clustered together, and one dies, the data on that one would not be available to you until it was restored.
I wasn't too keen on that idea. I was thinking more like striping across different nodes and restoring the missing data from checksums of the data that it had, but he was saying that that data would just be missing.
Not sounding like a viable solution to me. I'll keep my Synology's.
This means you don't want what they sell.
-
Ideally, if your planning on going down the VTL route, it would be great if you were to look into the direction of StarWind. They have recently presented their solution at the VeeamOn conference, after swinging by their stand, I've come to find out that its a pretty viable solution for the creation and storage of backups as tapes on S3 or Azure.
-
Ideally, if your planning on going down the VTL route, it would be great if you were to look into the direction of StarWind. They have recently presented their solution at the VeeamOn conference, after swinging by their stand, I've come to find out that its a pretty viable solution for the creation and storage of backups as tapes on S3 or Azure.
Yeah, it is very cool stuff.
-
Anyone seen ExaGrid pricing?
