One of the most interesting meetings that I attended yesterday, alongside VMworld, was the Tech Field Day. This event is not like a traditional TFD, this is a much more interactive session, a roundtable, to discuss a topic presented by a vendor.
Yesterday, in a room packed with bloggers from US and EU, two companies presented their ideas: Asigra and Infinio.

Asigra

First of all, I admit that I don’t know neither Asigra nor its products very good… and this session didn’t help either.
In fact we spent a lot of time talking about the licensing model of the product: on the contrary of many of its competitors Asigra charges on a recovery basis instead of a capacity of backed up Data. That’s interesting indeed and goes in a direction that most customers could love. The licensing mechanism is also clever enough to cover peaks, exceptions and recoveries for DR tests.

If you look at how data grow and how you really access them, you can easily find that most of those data are totally cold and never accessed. You probably still back them up every day, but the probability that you access and/or accidentally modify/delete/restore them back is very low. In this case this licensing model perfectly fits your needs.
On the other hand, from my point of view, if your company is experiencing such a growth of its (unstructured) data you should not rely on traditional backup mechanisms anymore. Backup costs too much, especially if you have a very long retention and cold data to mange. You should think about something different, like an archiving object storage solution (and this, and this) that can help you to avoid most of the OPEX to manage all your unstructured data.
I raised this point during the panel and the answer was that I’m not wrong but many companies are stuck to their backup systems and it’s very hard to impose/propose such a change in the mentality of many IT people…

Infinio

The second roundtable was sponsored by Infinio. I know Infinio through a Tech Field Day session that I attended on line a few months ago. The product is intriguing, to say the least, it’s a distributed software caching mechanism for VMware. It accelerates NFS datastores only though.
The product is in beta (it was released yesterday). You can download it from the company’s web site and it can be installed with a few clicks on you ESXi servers without service disruption… there is also a nice non-disruptive uninstall option.
The software installs a virtual appliance that allocates a small amount of RAM (8GB at the moment) and it immediately starts to accelerate storage traffic! Obviously, it uses a small amount of server’s CPU but that is acceptable, especially because the license is very cheap (499$ per socket, if I remember well).
Adding to the cost of the license, and the cost of the allocated resources on the server (CPU and RAM) the product remains very cheap, probably one of the cheapest on the market.
The product looks seamlessly integrated with vSphere and you can easily manage the tunables from an easy to use interface.

The only concern that I have about Infinio is about its maturity at the moment (it’s still in beta and it only supports NFS) and the risk that they will be soon acquired. In fact I like the product but it looks much more like a feature of broader solution than a product itself.

Disclaimer: I was invited to this meeting by Gestalt IT and they paid for 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.