Name: Evan Touger | Prowess Benchmark Testing Results for AMD EPYC on Dell PowerEdge
Uploaded: 2022-12-14T00:26:37.000Z
Duration: 17 min 40 s

More from test92226

03:57

18:05

Info

Evan Touger | Prowess Benchmark Testing Results for AMD EPYC on Dell PowerEdge

thank you [Music] welcome to thecube's continuing coverage of amd's fourth generation epic launch I've got a special guest with me today from prowess Consulting his name is Evan Taggart he's a senior technical writer with prowess Evan welcome hi great to be here thanks so tell us a little bit about prowess what does prowess do yeah we're a consulting firm we've uh been around for quite a few years based in Bellevue Washington and uh you know we do quite a few projects with folks from Dell to a lot of other companies and uh you know dive in we have Engineers writers production folks so pretty much you know end-to-end work doing research testing and writing and uh diving into different technical technical topics so you in this case what we're going to be talking about is uh some some validation studies that you've done looking at Dell poweredge servers that happen to be integrating in 4th gen epic uh processors from AMD um what were the specific workloads that that you were focused on in this study yeah this particular one was uh honing in on virtualization right um you know obviously it's it's pretty much ubiquitous in the industry everybody uh works with virtualization in one way or another so just getting Optimal Performance uh for virtualization was critical uh or is critical for most businesses so we just wanted to look a little deeper into you know how do companies evaluate that uh what are they going to use to make their determination for uh virtualization performance as it relates to their workloads um so that led us to this study where we looked at some benchmarks and and then went a little deeper under the hood to see what um what led to the results that we saw from those benchmarks so so when you say virtualization does that include uh virtual desktop infrastructure or are we just talking about virtual machines in general uh no it can include both we looked at VMS um you know thinking in terms of what about database performance when you're working in VMS um uh all the way through to vdi and companies like healthcare organizations and so forth where it's common to roll out lots of virtual desktops and and uh performance is critical there as well okay you you alluded to sort of looking under the covers to see you know where these performance results were coming from um I I assume what you're referencing is the idea that it's not just all about the CPU when you talk about a system uh am I correct in that assumption and yeah absolutely what can you tell us well you know for companies evaluating there's there's quite a bit to consider obviously right so they're looking at not just raw performance but Power performance um so you know that was part of it and and then what makes up that um those factors right so certainly CPU is critical to that but then other things come into play like the raid controllers um so we looked a little bit there and then uh networking of course can be critical uh for um configurations that are relying on on good performance on their networks both in terms of bandwidth and and just reducing latency overall so interconnects as well would be a big part of that so with with um pcie Gen 5 or 5-0 uh pick your moniker right um you know I in in this in the infrastructure game we're often we're often playing a game of whack-a-mole looking for the bottlenecks you know chasing the bottlenecks uh pcie5 opens up a lot of bandwidth from memory and things like raid controllers and Knicks um I mean does the bottleneck now just our imagination Evan have we reached a point where there are a bottlenecks what you know what were you able to stress to a point where it was saturated if anything well first of all we didn't say we looked at uh industry benchmarks and okay and we were examining in particular to see where world records were set um and so we uncovered a few specific servers poweredge servers um that were pretty key there or had a lot of um we're leading in the category in a lot of areas so that's what led us to then okay well why is that what's what's in these servers and what's responsible for that so in a lot of cases they uh we saw these results even with um you know uh Gen 4 pcie Gen4 so uh there were situations where clearly there was benefit from uh faster interconnects and especially nvme for uh raid um you know for supporting nvme and ssds but all of that just leads you to uh the understanding that it means it can only get better right so going from Gen 4 to if you're seeing great results in on Gen 4 then Gen 5 is probably going to be you know blow that away and in this case it should be even better in this case Gen 5 you're referencing pcie pcie right that's right and then the same thing with epic actually holds true some of the records we saw records set for both uh third and fourth gen so with Epix so the same thing there anywhere there's a record set on the third gen you know makes us really we're really looking forward to going back and seeing over the next few months uh which of those records fall and are broken by newer generation uh versions of these servers once they actually read to the newer generation processors um you know based on on what we're seeing for the for uh what this processors can do not only in sorry not only in terms of raw performance but as I mentioned before the power performance because they're they're very efficient and that's a really critical consideration right um I don't think you can overstate that for companies who are looking at um you know have to consider uh expenditures and power and Cooling and meeting sustainability goals and so forth so so that was really an important uh category in terms of what we looked at was that power performance not just raw performance yeah I want to get back to that uh that that's that's a really good point um we should probably give credit where credit is due which Dell poweredge servers are we talking about that were tested and what did those interconnect components look like from uh um yeah so we focused primarily on a couple benchmarks that seemed most important for real world performance results for virtualization uh TPC XV and uh vmmark 3.x the TPC XB that's where we saw uh poweredge r7525 r7515 they both had top scores in different categories there um that Benchmark is great for looking um at database workloads in particular right running a virtual virtualization settings and then the VM mark 3.x was critical we saw good good results there for the 7525 and the r7515 as well as the r6525 um in that one and and that included um sorry just checking notes to see what yeah no no no no it's that um included results for uh uh Power performance um as I mentioned earlier that's where we could see that so we kind of uh we saw this in a range of servers that included both Surge and AMD epic and uh newer 4th gen as well as I mentioned um the rate controllers were critical in the GPC XD I don't think that came into play in in the VM Mark tests but but they were definitely part of the TPC XV benchmarks so that's where the brake controls would make a difference right and uh and those tests I think they're using perk 11 so you know the newer perc 12 uh controllers there uh so again we'd expect to see continued you know gains uh in newer benchmarks that's what we'll be looking for over the next several months um yeah so I think uh if I've got my Dell nomenclature down performance no no power Edge RAID controller is that right exactly yeah there you go with broadcom uh you know powered by broadcom that's right there you go yeah so isn't isn't the Dell naming scheme there perk yeah yeah exactly exactly back to your comment about power so you've you've had a chance to take a pretty deep look at the latest stuff coming out you're confident that uh because some of these servers are going to be more expensive than previous generation now a server is not a server it's not a server but some are Awakening to the idea that there might be some sticker shock um you're confident that the bang for your buck the bang for your kilowatt hour is actually going to be beneficial we're actually making things better faster stronger cheaper for more energy efficient we're continuing on that curve that's what I would expect to see right I mean of course can't speak to surprisingly without knowing you know where the dollars are going to land on the servers but but uh I I would expect to see that because you're you're getting gains in a couple of ways I mean one if the performance increases to the point where you can run more VMS right get more performance out of your VMS and run more total VMS or more bdis then there's obviously a good um you know payback on your investment there and then as we were discussing earlier just the the power performance ratio right so if you're bringing down your your power and cooling costs if these machines are just more efficient overall then you should see some gains there as well so um you know I think the key is looking at What's the total cost of ownership over you know the standard like a three year period or something and and what you're going to get out of it for your number of sessions the performance for the sessions and the overall efficiency of the machines so so just just to be clear with these Dell poweredge servers you were able to validate world record performance but this is if you if you look at CPU architecture pcie bus architecture memory uh you know the class of memory the class of rate controller the class of Nick yeah those those were not all state-of-the-art in terms of at least what has been recently announced correct because that's right pcie 4-0 so to your point world records with that with that you've got next-gen raid controllers coming out next and Nick's coming out if the if the motherboard was pcie5 with commensurate memory all of those things are getting better exactly right I mean you're you're really you're just eliminating bandwidth constraints latency constraints you know all of that should should be improved um nvme um you know just collectively all these things just open the doors um you know letting more more bandwidth through reducing all the latency those are those are all pieces of the puzzle right that come together and and it's all about finding the weakest link and eliminating it and I think we're reaching the point where we're we're removing the biggest constraints from the from the systems okay so so I guess is it fair to summarize to say that um with this infrastructure that you tested you were able to set World Records this during this year I mean over the next several months things are just going to get faster and faster and faster and faster that's what I would anticipate exactly right if they're setting world records with these machines before some of the components are are you know the absolute uh latest it seems to me we're gonna just see a continuing Trend there and and more and more records should fall so I'm really looking forward to seeing how that goes uh because it's already good and I think the return on investment is pretty good there so I I think it's it's only going to get better as as these roll out so let me ask you a question that's a little bit uh a little bit off topic uh okay kind of um you know we see these gains you know we're all familiar with Moore's Law we're familiar with um with uh you know the advancements in memory and and bus architecture and everything else um we just covered supercompute uh 2022 in Dallas a couple of weeks ago and it was fascinating talking to talking to people about advances in AI uh that will be possible with new architectures no most of these supercomputers that are running right now are n minus one or n minus two um infrastructure you know they're they're PCI three right and maybe two generations of processors old because you know just throw out a hundred thousand CPU super Computing environment every 18 months it doesn't work that way exactly do you have an opinion on this question of the qualitative versus quantitative increase in Computing moving forward um and do you I mean do you think that this new stuff that you're starting to do tests on is going to power a fundamental shift in Computing or is it just going to be more consolidation better power consumption do you think there's an inflection point coming that's a that's a great question that's that's a hard one to answer I mean it's it's probably a little bit of both because certainly there will be better consolidation right um but I think that you know the systems uh it works both ways it just allows you to do more with less right um and you can go either direction you can you can do what you're doing now on on fewer machines um you know and get get better value for it um or reduce your footprint or you can go the other way and say wow this lets us add more machines into the mix and take our our level of performance from here to here right um so it just depends on what your focus is certainly with with areas like you know HPC and Ai and ml um having the ability to expand what you already are capable of by adding more machines that could do more is gonna gonna be your your main uh concern but if you're more like a small to medium-sized business and the opportunity to to do what you were doing on on a much smaller footprint and for lower costs that's really your your goal right um so I think think you can use this in either direction and it should should pay back in a lot of dividends yeah thanks for your thoughts it's an interesting subject moving forward um you know uh sometimes it's easy to get lost in the minutia of the bits and bites and Bobs of all the components we're studying but they're powering something that uh that's going to affect effect effectively all of humanity as we move forward um what else so what else what else do we need to consider when it comes to um what you've what you've just validated in the uh virtualization testing anything else we anything we left out I think we had all the key points for most of them it's you know really uh it's just keeping in mind that it's all about the the full system the components not you know the processor is obviously a key but um just removing uh blockages right uh freeing up getting rid of latency improving bandwidth all these things come to play and then the power performance as I said I I know I keep coming back to that but you know we just in a lot of what we work on we just see that businesses that's a really big concern for businesses um and uh finding efficiency right and especially in an age of constrained budgets that's that's a big deal so it's really important to have that power performance ratio and that's one of the key things we saw that stood out to us in in some of these benchmarks so well it's a big deal it's all good yeah I I live in California and I know exactly how much I pay for a kilowatt hour of electricity I bet yeah don't even know yeah I totally understand the power constraint question yeah it's not going to get better so yeah anything you can do there right yeah well Evan this has been great uh thanks for sharing the results that prowess has come up with third party validation that you know even without the latest and greatest components in all categories Dell Powerade servers are able to Set uh set world records and I anticipate that those World Records will be broken in 2023 and I expect that prowess will be part of that process so thanks thanks for that uh for the rest of us here at the cube I want to thank you for joining us stay tuned for continuing coverage of amd's fourth generation epic launch uh for myself and for Evan tauger thanks so much for joining us [Music]

Evan Touger | Prowess Benchmark Testing Results for AMD EPYC on Dell PowerEdge