Last week I attended Storage Field Day 5 and the most controversial session was the one with EMC.
If you didn’t get the chance to take a look at the videos recorded during that session you can find several links on Tech Filed Day YouTube channel (but I know, 4 hours is too long for anyone so I hope this short post will help!)
In this blog I’ll try to recap what I’ve learned (the good part) and what I haven’t (the bad part… and it’s not my fault).
The second part of the EMC presentation was dedicated to ViPR.
ViPR is a fascinating tool aimed at automating storage provisioning and I have to admit that this was the first real demo I had the chance to attend since the product launch.
First of all, I don’t know why EMC calls it “software-defined storage”, which it isn’t if compared to solutions like ScaleIO or VSAN (just to stay in the EMC realm) but in any case, it looked pretty interesting.
Even though I still have many unanswered questions, the product showed many intriguing aspects. (it’s a shame we didn’t spend more time on it!)
In practice, ViPR is capable of discovering and managing different storage components (ranging from Arrays to Filers and networking switches) which are used to build configurable pools of storage.
Once you have all these resources available, ViPR presents them through a self service portal and the user can request storage resources without having to configure each single component. Brilliant!
Yes, I’m oversimplifying, but you should really look at the demo to understand how easy it is to go through the provisioning process and get what you need. The potential is huge and a provisioning process that usually takes days (if not weeks) could be completed in hours (if not minutes!).
As mentioned above, I would like to know more about what happens behind the scenes, the actual limitations of the product, how smart it really is and so on. And at the same time, as far as I know, ViPR requires a lot of refinements before considering it for complex environments.
At the moment, for example, ViPR supports only EMC and NetApp storage while integration with other vendors will come later… and then I’m not sure whether the product will be able to maintain the same level of quality and simplicity when it will have to configure other vendor arrays. Only time will tell.
Another aspect of ViPR is the VM that provides data services (object storage access through S3 and Swift APIs). This is probably the part that makes ViPR more “software-defined”, but I’m afraid it doesn’t convince me. It’s free (so, you don’t pay for it) but it has its limits in value.
The VM (or probably the VMs) accessing a file-based backend with an object storage API at the front-end could be good for development and test environments, but nothing more.
I won’t spend too much time talking about the first part of the session. You can find videos, comments and opinions on the the web concerning that and I won’t be contributing to this useless discussion.
The problem is that we, as TFD delegates, are not interested in marketing stuff and EMC spent too much time talking “marketish”, a language that none of us understand.
The bloggers in the room were there to understand and learn and, even when we moved away from the pure marketish, we were disappointed by XtremIO. (I know, XtremIO’s CTO hosted the presentation but it was very hard to get answers from him).
that’s a shame, especially because the best presentation I’ve attended so far on XtremIO since their product launch, is the one from Solidfire’s CEO (Dave Wright) just a few hours after the meeting with EMC… and it didn’t look that bad.
Why it is important
ViPR is an interesting piece of technology and it deserves a certain level of attention. It’s immature, but it has the potential to become a powerful tool.
It is also interesting to note that ViPR is available as a free download for evaluation purposes.
My opinion on XtremIO is still on hold. It is immature, it has some potential flaws, but it also has some really good cards to play and EMC can theoretically spend more money than anyone else for its development.
Last but not least, EMC spends too much time (and money?) on marketing…..nothing new really. 😉
Disclaimer: I was invited to this meeting by Gestalt IT and they paid for travel and accommodation, I have not been compensated for my time and am not obliged to blog. Furthermore, the content is not reviewed, approved or published by any other person than the Juku team.