Don't drink the kool aid

When you put Pure in your data center it becomes a black box. Issues will happen with the Pure and you will receive no notifications from it. We've had controller failures, which you would think you'd want to be notified of, but you don't get those notifications. We've had numerous issues where we aren't made aware of issues with Pure. You just don't have the visibility into what's going on with a Pure that you would have with say an EMC array where you can look at every possible metric. Pure deems what the customer should be notified of and what performance metrics they can see which is not that detailed. We have discussed this with Pure and they indicate we are one of the only customers that would like to see this which seems to me to be a little odd. Who would not want to know what is going on with there storage environment?

Support does not communicate that well and they come out with code updates way to frequently. You can subscribe to their feed that lets you know when a code upgrade comes out. The release notes for the code upgrade will be very generic and you have no idea if they are applicable in your environment so you have no idea the sense of urgency to apply these fixes. They've been coming out pretty frequently also. We'll test an upgrade on one of our arrays for a few weeks to a month and by the time we've finished testing and burning in they will have come out with a newer version. Then you'll contact support to have them apply the version you've been testing on your other arrays and you'll have to be very clear with them on which version and why you want that version applied. There support representatives don't communicate that well either. We experienced a significant issue that caused a significant outage and the support representatives were not involved, even though they were aware of the issue, until we expressed our displeasure with Pure.

We experienced a significant bug with Active Cluster which brought a large portion of our production environment down. First response from Pure was not very timely nor effective in addressing the issue. Within a couple of weeks we experienced the issue again. At that point we got everything off of Active Cluster. After several weeks of pointing fingers at VMware and our configuration we were able to get more senior people at Pure engaged. It took almost 5 months for the issue to be identified and it was a Pure bug with Active Cluster. We have not put anything back onto Active Cluster as of yet and it's 7 months after the issue. Pure would have not stepped up to help us out without us letting Pure know how we feel about the product.

They give you great pricing to get in the door, but then when you have to grow you'll get sticker shock.