Hello, welcome to the 3rd episode of myth busters. Today will be taking on the myth of the misperception that vSAN is not ready for enterprise applications or business
critical applications and with me today is Fahid Faraduni. He is on our product marketing team focused on “go to market”. My name is Ernesto Cheena I’m director of the of the storage and availability buisiness unit So let’s get right into it Fahid Why don’t you help me address the question Are we ready for the enterprise with vSAN Yeah, absolutely First thing is that let’s look at our customer adoption and how much revenue we started generating by now we have more than 8,000 customers your customer adoption of vSAN has been accelerating very fast and a lot of those customers are using vSAN for business qualifications second aspect is all of the enhancements we have done to the product In terms of features. So we just released our 7th edition of the product and by now we have added deduplication, compression, & ensure coding. We are the first HCI vendor that has completed software-defined encryption. I have also done a lot of work around stretched clusters and an ability to have local protection as well. Putting these two together in terms of customer adoption – the fact that the revenue around that is 300 million this year, which we publicly announced.
This provides a great opportunity for our partners. The second is on features – we are traditional infrastructure and in many ways we are way ahead of them and this is why our partners should start proactively selling vSAN. Perfect, I think that the customer list as well as the fact that we’re on the 7th generation is impressive right? So this should make partners feel more comfortable. The second piece that they talk about is performance. Are we ready? Does our solution have the performance that is needed for enterprise application? Yes, absolutely. As a matter of fact, from the beginning when we release the product, one of the major focus areas for us has been performance. If you look at our performance numbers now we are publishing 100,00-150,000 IOPs per node. If a customer has a 6-node cluster they can easily get 400,000-500,000 I/O per second, which is way beyond what most of our customers need. In terms of latency we are providing sub-millisecond latency on all-flash configuration, that helps our customers a lot as well. Yes, those are actually impressive numbers. That actually should cover most applications. Not to put you on the spot, but there is one question we get a lot – it’s specifically around rebuilds and recyncs. So can you hit that one a little because I know that comes up in a lot of partner conversations. As I mentioned, we are at the 7th generation of our software and we’ve done a lot of enhancements addressing the resync issues. One is avoiding all of those together. The second one is making them very very intelligent. For example, initially, customers could go pick any hardware components they want. The problem with that was then we would end up in a configuration that the customer had maybe a controller that was not suitable for that platform.
Since then, we have come up with our ReadyLabs, which is a joint initiative between our engineering team and and our partners like HP, Dell, Cisco and others that we certify. and since it’s certified, it’s going to help them with their performance and also it’s going to help avoiding those failures that a lot of the time was related to those controllers, for example. The other thing that we have been doing is predicting drive failures. If we realize that the drive is failing, then we proactively evacuate the data from the drive so that it’s not going to be necessary for us to do the rebuild once the drive fails. The second aspect that I mentioned was making it very very intelligent. We want it very very intelligent so that, for example, if the controller driver has an issue to our cloud connected portal we notify the customers that the driver is outdated so that they can update that and avoid potential issues that can happen. So, putting it all together by doing the avoidance and making it very intelligent we have addressed most of those concerns. One last thing I want to mention is all of the stuff that I talked about – optimization and everything that we have done applies to both hybrid and all-flash but the fact that 60-70% of our customers are adopting all-flash helps with a lot of that situation a lot as well. Ya i know flash solves a lot of administrative issues and performance and all that. So, there are actually quite a few elements to being enterprise ready support being a critical one for customers. Tell me a little bit of how we handle support that meets the requirements from an enterprise perspective. Ya absolutely! If you look at support, there are several aspects into the support itself, right. The first thing is the product itself. and we’ve done a lot of enhancements to the product so that it’s going to enable the customer to use the product with out filing a support request. Let me give you an example. For example, initially when we release the product we record multicast on a networking site which adds a lot of complexity to our customer’s network configuration and we used to get a lot of support requests based on that. In 6.6 release that just came out, we removed the requirement for multicast anymore. It’s all going to be unicast, so that is going to help address all of those issues. So we have done a lot of optimization from the features perspective. The second aspect is the operations management. Some of the partners who looked at the product maybe in 2014 or early 2015 may have noticed that we used to record a lot of commandment operations. Now we have done a lot of operations built into vCenter UI, so you don’t need to use commandment at all. So it helps customers from the operations management and troubleshooting. The third thing is, on our side we have done major enhancements and investments on our support resources. Putting it all together with all of the product enhancements we’ve done and all of the enhancements we have made in our support personal our c-sat has gone from 85% in Q116 to now more than 95% and we are going to see more and more investment in that area and that is going to make our customers happier. Perfect. We’ve covered a lot of ground here. Thanks a lot Fahid on the support side We’ve covered performance, we’ve covered features. I think we’ve covered all of the elements someone would be concerned with in regards to being enterprise ready, so i think we busted that myth! Thank you very much.