 Can you see it? Yeah, so so we're here at the open server summit. So who are you? Anil Vasudeva with the president of IMAX Research, which is an industry analyst company We do both research and consulting for the IT industry specifically now very focused on software-defined data centers, software-defined storage and software-defined networks But that is coming on very strong in the last two years and it'll catch on fire It's it may be an inflection point for this IT industry and the reason is because it really brings the Operating cost down and and that's what all the data centers are looking for so it's kind of like a required way of doing things to Make things work when there's so much more demand for stuff and the power there's a limit, right? Right I mean let me give you a metaphor a metaphor would be a body is a hardware and Then depending upon your function that you're trying to go if you go in the evening You put a dress you are going in the morning and doing some casual you put a jeans and you go to office And you could put a jacket so those in effect is the at the end of the day Since the hardware is not that we are putting the money in you are creating different personality for different Functions, so some of the functions are you know enterprise functions, which are very heavy-duty and required the robustness as well as require Shall we say you know unforgiving It cannot break down and all that whereas on the other hand you have Videos and movies and all that those are not as needed to be robust and also need very extremely high Bandwidth or you know high speed of data transmission. So those are two different markets So that means if we can use the same computers But let the software make the personalities for very high bandwidth oriented videos Also the same software now switch over all these components join them together to do the enterprise heavy-duty work So now we have reduced the cost with the same computers. So that is the essence of software defined Future of data centers and you were talking about pooling of CPU memory different places. What is that? Right? So the pooling pooling means is look at this way I'm also going to start with the metaphor so spreading you have these jelly beans the yellow color and the blue color and this and So if you want a certain application Running and you need four of blue blue blue Jelly beans and two of these and four of these this is the equivalent metaphor of The kind of CPU you need and the kind of memory you need and the kind of storage So and and the network so because every computer runs on these fundamentals One is the CPUs the other one is the memory and the third one is IO, which is either network or Distributes and all that so now if you if you have a certain application It needs four of CPUs hundred of memories and needs of only two of IOS So you put them together and run the application and then give it back to the pool So the pool is whole bunch of IOS It's a whole bunch of memories and whole bunch of CPUs sitting there as pooled CPUs and pooled memory and pooled IO So you grab some of those depending upon what the application is needy and you run the application and give it back to The pool and that is the most and so the pool can be the most cost effective Why because it's the same models. It's the same jelly beans of one color another jelly beans of another color and jelly beans of third So when you want to build something you pick up two of these four of these depending upon what you need jar So the software defined and the pooling and these things sounds like new way of doing things Is that an opportunity for arm to come in and kind of like customized stuff? Okay, arm is one another incarnation of this whole So CPU it basically what it boils down to is that if you have Very general purpose low cost not not so much low power So you gang them whole bunch of these on CPUs together to create these functions as compared to having very individualized Very individualized very function oriented CPUs that Intel made in the past so and since the power is a very key function of Assembling a whole bunch of so think of this way if you can pull a whole bunch of arm processors and So you can carve them to make a Application to be run or a be application or see application. So that's why the success of it So you think it's gonna Succeed what do you think arm would have a very strong place in a whole bunch of early on not so much into the enterprise very high Demands but more so we're let's say for example, if you look at Facebook and all that they have Thousands and thousands of servers. So and they are all very dedicated to serving your Your information coming in from the web, so they fit in beautifully there And so the company like Facebook and the like e-bays and all that would be doing all these At the expense of some very dedicated high enterprise level Computers that Intel makes so a lot of change a lot of things are happening in this industry Well, the biggest change is not so much the arm versus Intel the biggest change is software So these are the elements that you assemble together and if you can do them on the fly on the fly then you are Serving those functions, then you give it back to the pool then next function comes you give it and serve that Function give it back to the pool. So it's a very new different ways of doing it And it all started with something called virtualization that was the first concept coming out of Stanford And is it Stanford again doing the software defined networking? So the other guys over there They're their hands and many you know I mean if you go if you go to Stanford and look at these evening meet-ups and all that oh You'll be seeing a whole bunch of things coming in from there continuously. It's a it's a It's a building ground for You know the next generation of innovators if you will