Some highlights from VMworld 2016

vmworld2016With Virtual SAN gaining huge momentum, I had a pretty busy VMworld 2016 in Las Vegas, and I didn’t have too much free time this year to immerse myself in the Solutions Exchange. However there were a few folks that I did want to catch up with as I had heard that they have some cool new features added to their product sets. So I made a bee-line to check them out. Let me know if you found anything else interesting. I might have a little more time in VMworld Barcelona to check out the solutions exchange and see what else is going on.


One of the first meetings I had was with Arun Agarwal of Infinio who showed me their new implementation of a cache accelerator using VAIO, the vSphere APIs for I/O Filters. Basically, using SPBM, you can associate a policy with a VM and utilize cache acceleration (basically directing I/O to a faster medium such as local flash, and later destaging to persistent storage such as your array). This all plays into VMware’s Software Defined Storage vision where data services can be attached to a VM through a policy setting. Arun talked us through their deployment and configuration steps, and then showed us the numbers they were able to achieve. Very impressive. Right now, the filter only allows for read cache acceleration, there is no acceleration on the writes. However Infinio are also able to leverage memory as well as flash devices, which accounts for the great figures that they are able to achieve. Worth checking out if you are in the market for a cache acceleration product.


cohesityCohesity is one of those companies that I have been watching for a while. I have written about them before, after I first met them at VMworld 2015. The company is the brainchild of Mohit Aron, who was one of the co-founders of Nutanix. I also wrote about their 2.0 release earlier this year. Their focus is to take care of all your secondary storage needs. I was given an overview of the next version of their product (DataProtect 3.0) by their VP of Marketing, Patrick Rogers and by Sunil Moolchandani, who heads up their Product Management team. The really interesting part for me was their integration with Pure Storage. Cohesity have the ability to replicate any snapshots that are taken on the Pure array, store them on their secondary storage platform, and then inform the array that it no longer needs to maintain these snapshots. At any point that you need to recover one of these snapshots, they have the ability to instantiate it back on the Pure Storage array. This means that the “primary” storage can be relieved of the task of maintaining these snaps. Very slick, and something that they might possibly extend to other storage arrays if there is a demand.


Last, but not least, I got an opportunity to catch up with my friends at Rubrik, namely Julia Lee (head of marketing) and Chris Wahl (Chief Evangelist). Rubrik are another company that I follow with great interest, and I have also written about them in the past. I first met them at our partner exchange last year, and earlier this year Chris and I wrote a joint VSAN/Rubrik white paper. What I really like about Rubrik is, just like VSAN, they have a very nice Software Defined Storage offering where the availability of your data is policy and/or SLA driven. Have a peak at the whitepaper if you wish to learn more. I caught up with Chris and Julia, and got a demo of the new features that they have recently introduced in Rubrik Firefly 3.0. The main items of interest is support for backing up of physical workloads, such as SQL server and Linux machines (a big ask from their customers). The other big-ticket item is the extension of “cloud” use cases. First, you can use the “cloud” for a backup destination (think AWS S3 or Azure). Now, depending on the policy/SLA, you can back up locally, or backup remotely to the cloud. The other new feature of interest was the introduction of an edge appliance, which is essentially a virtualized version of Rubrik’s appliance. This is ideal for remote office/branch offices (ROBO) with just a few VMs or servers, as it will allow backup data to be sent to your main DC running physical Rubrik appliances, while maintaining some local cached data in the ROBO for fast restores.

For those of you waiting to attend VMworld 2016 in Barcelona later this year, I would strongly recommend checking out these companies in the Solutions Exchange.