Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Copper enables the ARM server ecosystem

dell.com

Post reply on HN
Print view View original

61–70 of 94 posts

Re: Copper enables the ARM server ecosystem

#61

Post by Sssnake »

voidlogic wrote: post #33

I think x86/amd64 virtualization has made ARM a much less compelling option for servers than what it would have been a few years ago. I'm looking forward to benchmarks of new ARM server CPU (esp. AMDs), but in the past comparing scale out ARM boxes and tradtional servers in the same space has: 1. Been much more more in favor x86/amd64 under low load (much lower latency for users) 2. Been pretty similar under very hig…

>that is 32 logical cores at 3.0 GHz

Just having 32 logical cores doesn't mean much for performance. Hyperthreading is not magical, dual quad core cpus is still 8 cores, with potentially some minor performance gains from the hyperthreading depending on the application.

Re: Copper enables the ARM server ecosystem

#62

Post by mjg59 »

notacoward wrote: post #44
ck2 wrote: post #32

Are these really better per watt at serving content? I'm all for x86/intel competition but I don't see this as real savings yet?

Looks like they'd be a lot better per watt. http://armservers.com/2012/06/18/apache-benchmarks-for-calxe... This is hardly surprising to me. I worked on similar low-power high-density systems at SiCortex for a couple of years. High clock rates and big caches do improve performance, but they increase power consumption even more, so for a parallelizable workload a larger number of "wimpier" processors can do more work…

Those figures are pretty close to being lies. A quad-core 3.3GHz Xeon that's only serving 6950 requests per second isn't going to be at anywhere near 100% utilisation, which means there's no way it's running at anywhere near its TDP. Same for the RAM. The x86 isn't going to be anywhere near 5W, but it's going to be significantly lower than 102W. They're also ignoring the fixed disk and PSU overheads, shifting the power side of the ratio in favour of the Calxeda, and using a benchmark that's going to end up network limited is a solid way of reducing the performance advantage of the Xeon.

I don't doubt that ARM does have a better power/performance ratio, but it's nowhere near 15x. Using such obviously misleading statistics leaves me suspecting it's way closer than that.

Re: Copper enables the ARM server ecosystem

#63

Post by iyulaev »

dham wrote: post #28

Maybe I'm not opened minded enough, but I just don't believe the future is ARM on the desktop/server. I believe the future is x86 on the phone. I feel like it's less time until Intel gets the power correct as opposed to ARM getting the speed.

Bay Trail atom is already competitive with tablet-focused ARM. I've got a Samsung 500T that runs "real" windows. It weighs as much as an iPad and has ~10 hour run time, same as the iPad. The whole microarchitecture thing is less of an issue for power consumption and performance than most people think.

Re: Copper enables the ARM server ecosystem

#64

Post by jws »

ris wrote: post #36
ansible wrote: post #14

Earlier quoted context omitted.

If we could turn it into a NUMA machine with 48 CPUs, that would be a totally different story. This particular version seems to only support 1Gbit Ethernet, so I don't know how well that will work. Still, we're talking about 192 cores with 384 Gbytes of RAM, that's got to be useful for some kind of workloads. Can you get that kind of density with Intel or AMD?

> Still, we're talking about 192 cores with 384 Gbytes of RAM Seeing as you're going to be keeping 48 copies of the same OS in that memory whether you want to or not, you're going to need those 384G.

I just checked a random server: OS and all process images are using 119MB, *48 = 5.5GB. I think I'll do just fine with the remaining 378G.

Re: Copper enables the ARM server ecosystem

#65

Post by ams6110 »

lazyjones wrote: post #49
wtracy wrote: post #39

Earlier quoted context omitted.

> Quad socket x86 is still prohibitively expensive Is there an engineering reason for this, or is there just not enough volume for those boards to be economical?

I'd vouch for the 3rd possibility: the prices are artificially high because the customers who buy these configurations can afford to pay the premium.

This. Purchasers of the highest-end and/or costly experimental architectures tend to be research institutions or national labs that are using grant money.

Re: Copper enables the ARM server ecosystem

#66

Post by ams6110 »

happywolf wrote: post #54
venomsnake wrote: post #35

Earlier quoted context omitted.

It won't - it may evolve or change the PC market but it won't pressure it. People will need computing powerhouses and big monitors. My predictions are that we will hit cloud disillusionment soon and people are going to rediscover the benefits of home storage and GPU power.

The issue is CPU no longer is the bottleneck since a while ago. My three-year old laptop with a then high-end graphics card still serves me well. There of course are still some people who need powerful CPUs, but I guess most people will need more powerful GPU instead of CPU.

My old G4 PowerBook still serves me well for many things. Most people need a lot less hardware power than they think they do.

Re: Copper enables the ARM server ecosystem

#67

Post by zokier »

_delirium wrote: post #59
ghostdiver wrote: post #20

Earlier quoted context omitted.

This type of infrastructure is not insta win for many(most.. ?) of businesses. Even if the price will be low and the power consumption will make a difference, still there will be HUGE cost of rewriting software. Not all businesses need to write scalable software, because their current technology stack is just good enough and will always fit to one beefy machine, but on the other hand it will never fit on one ARM serv…

That really depends on what kind of software you're running. There's a wide range of portable C/C++ Unix stuff that compiles fine on ARM. If you run a distro like Debian or ubuntu, you can just install the armhf port and have all that stuff. Where it gets tricky is if it's something written for an environment that itself has no ARM port yet, such as the JVM.

I think its more of a question of (re-)architecting the software to perform well on bunch of wimpy cores vs normal brawny x86 ones.

Re: Copper enables the ARM server ecosystem

#68

Post by caw »

Sssnake wrote: post #61
voidlogic wrote: post #33

I think x86/amd64 virtualization has made ARM a much less compelling option for servers than what it would have been a few years ago. I'm looking forward to benchmarks of new ARM server CPU (esp. AMDs), but in the past comparing scale out ARM boxes and tradtional servers in the same space has: 1. Been much more more in favor x86/amd64 under low load (much lower latency for users) 2. Been pretty similar under very hig…

>that is 32 logical cores at 3.0 GHz Just having 32 logical cores doesn't mean much for performance. Hyperthreading is not magical, dual quad core cpus is still 8 cores, with potentially some minor performance gains from the hyperthreading depending on the application.

In my experience with HPC (i.e. 100% load on all cores), enabling Hyperthreading gives you higher job throughput at the cost of increasing the time to complete a single job. The throughput improvement is around 5-30% depending on the applications being run. The average I've seen is around 20.

Re: Copper enables the ARM server ecosystem

#69

Post by theatrus2 »

_delirium wrote: post #59
ghostdiver wrote: post #20

Earlier quoted context omitted.

This type of infrastructure is not insta win for many(most.. ?) of businesses. Even if the price will be low and the power consumption will make a difference, still there will be HUGE cost of rewriting software. Not all businesses need to write scalable software, because their current technology stack is just good enough and will always fit to one beefy machine, but on the other hand it will never fit on one ARM serv…

That really depends on what kind of software you're running. There's a wide range of portable C/C++ Unix stuff that compiles fine on ARM. If you run a distro like Debian or ubuntu, you can just install the armhf port and have all that stuff. Where it gets tricky is if it's something written for an environment that itself has no ARM port yet, such as the JVM.

OpenJDK Hotspot has a relatively mature ARM port. Due to "embedded" licensing it's not been supported from Oracle, but you can use it.

Re: Copper enables the ARM server ecosystem

#70

Post by kogir »

I'm honestly not interested in all the reasons you won't use these servers.

I doubt Dell designed these speculatively. One or more Dell customers likely requested them.

Who are these customers, and what is their use case? That's what's likely interesting.

Post reply on HN