 Hi, this is Stu Miniman with wikibond.org, Silicon Angle, on the ground with the Cube at Intel Developer Forum 2014. Joining me for this segment is Joram Novik, who is the CEO and founder of Maxta. Maxta, of course, the company that we've been keeping a close eye on, fits in a category that we've called ServerSan, which is a new generation of storage architecture that takes server compute, scale out architecture, hyper-converged solutions fit under this category, but it's a whole new way of building storage built on latest software, all the different hardware coming on. So Joram, we're here at IDF, and one of the reasons we want to talk to you is that Intel co-led the round of funding that you guys just had. Can you talk to us a little bit about Maxta's relationship with Intel? Yes, absolutely. So Maxta participated and actually led our latest financing round, but more importantly, Maxta is one of our development partners, and we have been working with Intel for quite some time on various products and features, and actually we are very happy to announce the first result of this relationship in yesterday's announcement of both our Maxta storage platform, which is a new way of hyper-converging compute storage and networking on the same set of server as well as our reference architecture called MaxDeploy, and this is all based on all of the latest and greatest products from Intel, so that's part of the cooperation that we are doing, and we are planning to deliver more and more, and basically enabling Maxta to maximize the promise of hyper-convergence by leveraging the latest and greatest coming of Intel as soon as Intel delivers those platforms. Alright, so we are talking about the grant release from Intel. Give us a little bit of insight, speeds and feeds, what is this latest generation of Intel allowed you to develop from the storage standpoint, the networking standpoint, what does it do to the whole platform? Yes, sure. So one of the challenges in the past for hyper-converges in general is that in the past you had application running on servers, and then you got storage systems that are basically processing, storing, and protecting data. By converging all those software functionality into one server, you need much more power and much more capability from the server, so Grantly is really a great platform for hyper-convergence because it is faster, it has more cores, so you can process more, and in our testing, and we have been testing Grantly, both the Xeon processor as well as the new SSD and the new 40 gigabit Ethernet, and we found that the performance increase is very significant. We found about 40% performance increase just by using this platform when compared to any prior generation. So Grantly basically provides all of the capability that are required to take hyper-convergence from very specific use case and making it a mainstream hyper-convergence that you can deploy all mission-critical applications on top of. Alright, so Joram, one of the discussions with Software Defined Storage is when it doesn't make sense to really have a standalone storage array say in all FlashArray versus do I have the horsepower and everything I need inside a single box, what's your lay of the land today? So I would say this, definitely all FlashArray have their place, and their place is usually for physical server. Whenever you have any application that requires mission-critical, you don't virtualize it in the first place. Virtualization does come with its own overhead. So in physical server, if you run your very intensive ERP, probably all FlashArray with physical server probably with high-end databases is the right way to go. For the rest of the application where virtualization made sense for the last 10 years ago by consolidating, by providing the simplicity, the availability and agility, hyper-convergence or server-send as Wikibon is calling it, is the right way to go. Because there is no reason for keeping and maintaining storage arrays if you can do everything within one platform. Why having two platforms if one platform can do it all? Alright, what about newer applications, what people might call the cloud mobile big data type applications that may or may not go virtualized, might even go with containers? So container is just the next step. So if you think about it, and this is really where my argument about physical versus virtual. So when we talk about hyper-convergence or server-send, we never said it's VMware, it's Hyper-V or it's KVM. It's really about once the server has some level of obstruction where the application are not running on the physical resources but are running on some level of obstruction, whether it's virtualization or container, I think hyper-convergence, server-send is the way to go. Because at that point in time, taking together the compute and the storage and optimizing the storage for whatever obstruction layer compute is running on is the right way to go. So containers and everything else that is going to come along is just going to continue to validate the hyper-convergence approach. Alright, so Yoram, last question I have is let's talk about customers. When you're engaging with customers, what's the real pain point that you're helping solve? What should somebody out there watching say, if this is what I've got today, I really need to call Maxsta up and it would be one of the solutions I should look at? So the main challenges that we are addressing with Maxsta is complexity, cost, and having availability and agility for storage. So what we provide to our customers is a much simpler approach of managing your infrastructure because you don't have to manage storage resources independently, you don't have to do all the mapping and all the complexity that come along with storage arrays, and at the same time, you can reduce the cost, whether it's OPEX or CAPEX. But one of the challenges going from traditional IT infrastructure into hyper-convergence, how do I maintain all the investment that I add in terms of availability, agility, data services, capacity optimization? So what Maxsta brings to the table is you don't have to compromise. You get all of those features and functions, while at the same time, you simplify and reduce the cost of your environment. And actually, some of the new features that we introduced yesterday are actually taking us one step further because we announced the ability to replicate the cost distance. So in the past, we could replicate within the cluster synchronously. Now we can do it across cluster. We introduce a new level of automation and we introduce more performance improvement. So if you take it all together, you get all the features and function, all of the capability of a storage array without requiring a storage array in the first place. Well, Joram, congratulations on all the updates to Maxsta and your product line, and thanks for taking time joining us. This has been Stu Miniman with theCUBE at Intel Developer Forum 2014 here in San Francisco. Thank you very much.