We use Amazon EBS for its different business applications, basic file shares, and database. You could call it a SAN. I'm not sure technologists would call it a SAN, but I do.
Amazon EBS has made our organization a little more efficient.
When it comes to valuable features, ease of use would be number one. Ease of access from anywhere type of thing.
I think performance can be improved. It's like feeling latency when you compare apples to apples as far as what you have in a SAN on-site, on-prem versus an EBS volume in the cloud. If you want to get rid of the latency, then that's where these costs start to really add up.
On the onset, it's cheaper to get started. It has every bit of capability and availability you would expect and even more than your own environment. If you're trying to get to the same or a higher level of performance, you have to bite the bullet and pay through the roof.
The ability to administer it, use it, set it up, install it, and operate it is all as easy or easier. You get the flexibility to change, snapshot, and all of the same things you could do with even more capabilities. But if you have to get the same or improve your performance, it gets exponentially more costly.
I have been using Amazon EBS for over three years.
As far as stability, availability, and durability, and all of that, it's basically just about perfect.
Amazon EBS is easy to scale. It's easier than on-prem.
The initial setup is pretty straightforward. If you've ever set up LUNs and block storage in the physical world, it's pretty much the same. Just setting up a partition and carving up the LUN, and that's about it.
Amazon EBS is expensive. What I want Amazon to do is lower the price. All the functionality that I could think of and would like to consume is already there. It's just that there's a subset of it that's not affordable to us yet.
We're still using it on-prem, and it's just a matter of migrating to the cloud. The way I look at it, there are about 27 advantages to migrating to the cloud. There are one or two disadvantages, and these disadvantages can be overcome with money.
It's one of those things where we're selective about what's getting moved by measuring it and characterizing it. Then we pick and choose what moves next. There are some tremendous advantages to moving up to the cloud, and that's the reason why we started this effort.
Making sure that the effort would work and would satisfy where we were going was step one. Then step two was taking, picking, and choosing which workloads to move and when to move them. Then deciding if there are workloads that would never move. I'm not sure I think that's the case yet, but it's just a matter of time and budget.
If you are thinking of getting Amazon EBS, I would advise you to characterize your priorities and incrementally measure the outcomes as you move the groupings of workloads over. Make sure that your predictions were the way it worked in reality. Then, measure the new reality, and verify with your users. Check whether the limitations and the disadvantages are acceptable to your users. If they aren't acceptable, then deliver the needed cost increments that would deliver upon that.
On a scale from one to ten, I would give Amazon EBS an eight.
