Personal computing discussed

Moderators: renee, mac_h8r1, Nemesis

 
DancinJack
Maximum Gerbil
Topic Author
Posts: 4489
Joined: Sat Nov 25, 2006 3:21 pm
Location: Kansas

Big release day for Intel

Tue Apr 02, 2019 12:19 pm

i7 6700K - Z170 - 16GiB DDR4 - GTX 1080 - 512GB SSD - 256GB SSD - 500GB SSD - 3TB HDD- 27" IPS G-sync - Win10 Pro x64 - Ubuntu/Mint x64 :: 2015 13" rMBP Sierra :: Canon EOS 80D/Sony RX100
 
Goty
Gerbil
Posts: 31
Joined: Wed Mar 04, 2009 7:41 am

Re: Big release day for Intel

Tue Apr 02, 2019 12:47 pm

Ouch @ 400W TDP for the 56 core Cascade Lake-AP part. The spec bump from the original 48 core announcement is kind of funny, but not entirely unexpected given the assessment from some OEMs that Cascade Lake-AP wasn't going to be a compelling halo part when compared to Rome. It might end up faster than the top Rome SKU, but at probably twice the TDP, is anyone really going to care?
 
DancinJack
Maximum Gerbil
Topic Author
Posts: 4489
Joined: Sat Nov 25, 2006 3:21 pm
Location: Kansas

Re: Big release day for Intel

Tue Apr 02, 2019 12:56 pm

Goty wrote:
Ouch @ 400W TDP for the 56 core Cascade Lake-AP part. The spec bump from the original 48 core announcement is kind of funny, but not entirely unexpected given the assessment from some OEMs that Cascade Lake-AP wasn't going to be a compelling halo part when compared to Rome. It might end up faster than the top Rome SKU, but at probably twice the TDP, is anyone really going to care?

This particular graf from AT sums it up pretty well I think.
Anandtech wrote:
While these new CPUs do not use a new microarchitecture compared to the first generation Skylake-based Xeon Scalable processors, Intel surprised most of the press at its Tech Day with the sheer number of improvements in other areas of Cascade Lake. Not only are there more hardware mitigations against Spectre and Meltdown than we expected, but we have Optane DC Persistent Memory support. The high-volume processors get a performance boost by having up to 25% extra cores, and every processor gets double the memory support (and faster memory, too). Using the latest manufacturing technologies allows for frequency improvements, which when combined with new AVX-512 modes shows some drastic increases in machine learning performance for those that can use them.

re: The 9200 series parts
Anandtech wrote:
Intel will be offering three parts, and where previously the company stated it would offer up to 48 cores in a single package, it will now offer up to 56, with the top TDP up to 400W. These processors are BGA only, and will only be sold as fundamental Intel server designs by Intel through the OEMs. The OEMs can optimize the design as they see fit, such as offering four blades in a 2U design or going between air and liquid cooling, however the motherboard/CPU configurations will be fixed by Intel.
i7 6700K - Z170 - 16GiB DDR4 - GTX 1080 - 512GB SSD - 256GB SSD - 500GB SSD - 3TB HDD- 27" IPS G-sync - Win10 Pro x64 - Ubuntu/Mint x64 :: 2015 13" rMBP Sierra :: Canon EOS 80D/Sony RX100
 
Goty
Gerbil
Posts: 31
Joined: Wed Mar 04, 2009 7:41 am

Re: Big release day for Intel

Tue Apr 02, 2019 1:05 pm

There's a lot of good information over at STH about the part as well: https://www.servethehome.com/intel-xeon ... -launched/

Notable in that article is the fact that the 9200 series parts (Cascade Lake AP) does NOT support Optane DC Persistent Memory, which doesn't seem like much of a loss to me if these parts are really only destined for HPC applications. It's funny that the author of the article twice mentions that these parts are not competitors to EPYC and then goes on to show the slide comparing the 9242 to the EPYC 7601. Makes you wonder why they didn't make the comparison using a 922x part, though. ;)
 
chuckula
Gold subscriber
Minister of Gerbil Affairs
Posts: 2106
Joined: Wed Jan 23, 2008 9:18 pm
Location: Probably where I don't belong.

Re: Big release day for Intel

Tue Apr 02, 2019 1:07 pm

A very big launch even if the HBM equipped Icelake Xeons will make more headlines when they launch next year. The Optane memory won't make Cinebench run faster but it will generate major performance increases in the databases that drive the majority of eneterprise sales. Any school kid can say moar coars (and Cascade Lake AP has them) but no competitor to Intel has an answer for Optane.

The most exotic thing is the DDR5 and PCIe 5.0 equipped FPGAs. Those will be interesting going forward to see how they fit into different product lines. They are definitely being pushed for 5G applications but I'd be curious at where else they might go.

As for Goty's usual ignorance, there are plenty of HPC customers out there for whom a 400 watt socket (not really different from a high end GPU) with 56 cores of AVX-512 is a very exciting product. Don't think for one second that even a 150 watt Epyc 2 with 64 cores (and I'm being overly generous to AMD) will win a power efficiency benchmark on the HPC and inferencing workloads that these parts will tackle. Not to mention that if a Rome processor can even boost its cores up to the 2.6 GHz base speed of the highest end AP chips it sure won't be running in your laptop either.

There's a reason that Intel's masive Fab capacity is booked solid for the foreseeable future while TSMC is hoping the smartphone market rebounds because they sure aren't supporting their line with Epyc orders.
4770K @ 4.7 GHz; 32GB DDR3-2133; Officially RX-560... that's right AMD you shills!; 512GB 840 Pro (2x); Fractal Define XL-R2; NZXT Kraken-X60
--Many thanks to the TR Forum for advice in getting it built.
 
DancinJack
Maximum Gerbil
Topic Author
Posts: 4489
Joined: Sat Nov 25, 2006 3:21 pm
Location: Kansas

Re: Big release day for Intel

Tue Apr 02, 2019 1:10 pm

That's not really surprising. That's how companies do marketing.
i7 6700K - Z170 - 16GiB DDR4 - GTX 1080 - 512GB SSD - 256GB SSD - 500GB SSD - 3TB HDD- 27" IPS G-sync - Win10 Pro x64 - Ubuntu/Mint x64 :: 2015 13" rMBP Sierra :: Canon EOS 80D/Sony RX100
 
Goty
Gerbil
Posts: 31
Joined: Wed Mar 04, 2009 7:41 am

Re: Big release day for Intel

Tue Apr 02, 2019 1:16 pm

chuckula wrote:
As for Goty's usual ignorance, there are plenty of HPC customers out there for whom a 400 watt socket (not really different from a high end GPU) with 56 cores of AVX-512 is a very exciting product. Don't think for one second that even a 150 watt Epyc 2 with 64 cores (and I'm being overly generous to AMD) will win a power efficiency benchmark on the HPC and inferencing workloads that these parts will tackle. Not to mention that if a Rome processor can even boost its cores up to the 2.6 GHz base speed of the highest end AP chips it sure won't be running in your laptop either.


viewtopic.php?f=32&t=11214
2.) Personal attacks are frowned upon. If you have to resort to such behavior, perhaps you belong someplace else. It is understandable that some discussions can get heated but try to maintain civility. If you cannot, you may be subject to censure or other disciplinary measures.


Just in case you weren't aware.

I'm sure there are HPC customers that value performance over all else, but performance per watt is still a very important metric in that space as well. Again, I seriously doubt Intel is going to be twice as fast as AMD for twice the power, even in AVX-512 workloads. AMD should make some significant strides in that arena with Rome with full AVX2 support, so it will certainly be interesting. *EDIT* Just browsing Intel's own benchmarks, in an AVX-512 workload (NAMD), Intel states around a 75% performance advantage with a 50% core count advantage vs an EPYC 7601. If AMD hits their targets for clockspeed and power (or even comes particularly close), that might be a wash at best.

Intel should certainly have the advantage in memory bandwidth constrained scenarios, though, coming in with 50% more memory channels to play with.
 
Glorious
Gold subscriber
Gerbilus Supremus
Posts: 11763
Joined: Tue Aug 27, 2002 6:35 pm

Re: Big release day for Intel

Tue Apr 02, 2019 2:34 pm

Goty wrote:
I'm sure there are HPC customers that value performance over all else, but performance per watt is still a very important metric in that space as well.


He's not saying otherwise, just that there are customers who welcome this sort of thing.

Because that's true.

And, yes, you appear to be ignorant of that.

Goty wrote:
Again, I seriously doubt Intel is going to be twice as fast as AMD for twice the power, even in AVX-512 workloads


So what?

Oh, *YOU* have made your own metric of relevancy here?

To quote another poster here: "is anyone really going to care?"
 
Goty
Gerbil
Posts: 31
Joined: Wed Mar 04, 2009 7:41 am

Re: Big release day for Intel

Tue Apr 02, 2019 2:36 pm

Glorious wrote:
Goty wrote:
I'm sure there are HPC customers that value performance over all else, but performance per watt is still a very important metric in that space as well.


He's not saying otherwise, just that there are customers who welcome this sort of thing.

Because that's true.

Goty wrote:
Again, I seriously doubt Intel is going to be twice as fast as AMD for twice the power, even in AVX-512 workloads


So what?

Oh, *YOU* have made your own metric of relevancy here?

To quote another poster here: "is anyone really going to care?"


Just trying to apply the same scrutiny to Intel's parts that AMD's seem to receive. I apologize if that's uncomfortable. Performance per watt is hardly my "own metric of relevancy," and the question of absolute performance is still up in the air in a variety of workloads where Intel has so far maintained a performance and value advantage.
Last edited by Goty on Tue Apr 02, 2019 2:41 pm, edited 1 time in total.
 
Glorious
Gold subscriber
Gerbilus Supremus
Posts: 11763
Joined: Tue Aug 27, 2002 6:35 pm

Re: Big release day for Intel

Tue Apr 02, 2019 2:40 pm

Goty wrote:
Just trying to apply the same scrutiny to Intel's parts that AMD's seem to receive. I apologize if that's uncomfortable.


It's not "scrutiny" to ignorantly declare that no one is interested in a SKU because it doesn't match your pre-conceived notions of market suitability.

Especially when those notions are *OPENLY* about "FAIR PLAY FOR AMD!" and the even more asinine "EVERYONE ELSE HAS A PROBLEM--- NOT ME!"

---

Yes. I'm uncomfortable. This is a cringe-post.

*SHUDDER*
 
Goty
Gerbil
Posts: 31
Joined: Wed Mar 04, 2009 7:41 am

Re: Big release day for Intel

Tue Apr 02, 2019 2:42 pm

Glorious wrote:
Goty wrote:
Just trying to apply the same scrutiny to Intel's parts that AMD's seem to receive. I apologize if that's uncomfortable.


It's not "scrutiny" to ignorantly declare that no one is interested in a SKU because it doesn't match your pre-conceived notions of market suitability.

Especially when those notions are *OPENLY* about "FAIR PLAY FOR AMD!" and the even more asinine "EVERYONE ELSE HAS A PROBLEM--- NOT ME!"

---

Yes. I'm uncomfortable. This is a cringe-post.

*SHUDDER*


I don't believe I declared anything. Maybe you didn't read my post very carefully?
 
Glorious
Gold subscriber
Gerbilus Supremus
Posts: 11763
Joined: Tue Aug 27, 2002 6:35 pm

Re: Big release day for Intel

Tue Apr 02, 2019 2:44 pm

Goty wrote:
Performance per watt is hardly my "own metric of relevancy," and the question of absolute performance is still up in the air in a variety of workloads where Intel has so far maintained a performance and value advantage


I'll address the ninja:

This isn't even remotely what your started with. You openly said that no one would even care about this part.

That's ridiculously untrue.

Your "defense" of this is that "someone's got to stick up for AMD!" and then, belatedly, "hey we really should be talking about something else entirely"

Hey. Champ. Start that way next time then, OK?

Cause otherwise your disingenuity is not just showing, but blinding.
 
Glorious
Gold subscriber
Gerbilus Supremus
Posts: 11763
Joined: Tue Aug 27, 2002 6:35 pm

Re: Big release day for Intel

Tue Apr 02, 2019 2:45 pm

Goty wrote:
I don't believe I declared anything. Maybe you didn't read my post very carefully?


Goty wrote:
is anyone really going to care?


If you don't understand basic rhetorical devices, namely in this case the literally EPONYMOUS one...

...maybe don't use them, then?
 
Goty
Gerbil
Posts: 31
Joined: Wed Mar 04, 2009 7:41 am

Re: Big release day for Intel

Tue Apr 02, 2019 2:50 pm

I suspect it's simply your own biases causing you to read far too much into my post. Do you care to contribute to the discussion or would you rather just continue your ad hominem attacks?
 
Krogoth
Gerbil Elder
Posts: 5754
Joined: Tue Apr 15, 2003 3:20 pm
Location: somewhere on Core Prime
Contact:

Re: Big release day for Intel

Tue Apr 02, 2019 2:53 pm

2019-2020 are going to be interesting in the HPC/enterprise world. The 800 pound gorilla got up from its laurels but still has to content with some teething issues (namely 10nm-7nm process in the foundries).
Last edited by Krogoth on Tue Apr 02, 2019 3:01 pm, edited 1 time in total.
Gigabyte Z390 AORUS-PRO Coffee Lake R 9700K, 2x8GiB of G.Skill DDR4-3600, Sapphire RX Vega 64, Corsair CX-750M V2 and Fractal Define R4 (W)
Ivy Bridge 3570K, 2x4GiB of G.Skill RIPSAW DDR3-1600, Gigabyte Z77X-UD3H, Corsair CX-750M V2, and PC-7B
 
Glorious
Gold subscriber
Gerbilus Supremus
Posts: 11763
Joined: Tue Aug 27, 2002 6:35 pm

Re: Big release day for Intel

Tue Apr 02, 2019 2:56 pm

Goty wrote:
I suspect it's simply your own biases causing you to read far too much into my post.


You openly said that your agenda here is to level things in regards to AMD, implicitly claiming the current "scrutiny" here is unfair:

Let me quote you:

Goty wrote:
Just trying to apply the same scrutiny to Intel's parts that AMD's seem to receive.


So what bias do *I* have?

Goty wrote:
Do you care to contribute to the discussion or would you rather just continue your ad hominem attacks?


I just told you to not use rhetorical terms you don't understand.

What did I say that was ad hominem?
 
Glorious
Gold subscriber
Gerbilus Supremus
Posts: 11763
Joined: Tue Aug 27, 2002 6:35 pm

Re: Big release day for Intel

Tue Apr 02, 2019 3:16 pm

To be clear, Goty, what is even your argument?

You say "ouch" at 400W but given that's it is basically two pieces of silicon strapped together that are individually ~200W, uh, OK? Yes, some customers want that sort of density, meanwhile, what are the other ramifications? Is there an argument here?

You state that Intel's unexpected 56 as opposed to 48 was "funny", but I have no absolutely idea why. What's the joke? "Intel over-delievers hahaha!" Uhhh...

You mention utterly unsourced rumors that "OEMs" have decided that it won't be competitive with AMD's Rome. I've had "experience" with such things, so I provisionally do not believe you. Meanwhile, stipulating that there is ANY basis in fact: So what? Rome, to my knowledge, does not have announced TDP. Thus they are likely speculating, just as you are clearly speculating. So what are we even arguing about? Nothing?

You imply that it "might" end up faster than Rome, but at twice the TDP, which is basically alleging that AMD is essentially twice as efficient than Intel when it comes to performance/power in this context, which doesn't seem tremendously likely (to say the least...). And it's based on ..what? TSMC 7nm versus Intel 14nm++ or whatever? That's murky and uncertain, and one of them is still a planned release, not an actual release.


----

This whole approach just reeks of AMD sour grapes. Which is completely bizarre, because why does that matter? Maybe AMD's new thing will come out and just wipe the floor with Intel. Good!

But until that actually happens, what is the point of this nonsense?
 
Goty
Gerbil
Posts: 31
Joined: Wed Mar 04, 2009 7:41 am

Re: Big release day for Intel

Tue Apr 02, 2019 3:29 pm

Glorious wrote:
Goty wrote:
I suspect it's simply your own biases causing you to read far too much into my post.


You openly said that your agenda here is to level things in regards to AMD, implicitly claiming the current "scrutiny" here is unfair:

Let me quote you:

Goty wrote:
Just trying to apply the same scrutiny to Intel's parts that AMD's seem to receive.


So what bias do *I* have?

Goty wrote:
Do you care to contribute to the discussion or would you rather just continue your ad hominem attacks?


I just told you to not use rhetorical terms you don't understand.

What did I say that was ad hominem?


Ad hominem (Latin for "to the person"), short for argumentum ad hominem, is a fallacious argumentative strategy whereby genuine discussion of the topic at hand is avoided by instead attacking the character, motive, or other attribute of the person making the argument, or persons associated with the argument, rather than attacking the substance of the argument itself.


If you can't get from point A to point B from that, I'm not sure how to proceed here.

Back OT, I find STH's value analysis of the second generation Xeon Scalable SKUs interesting when it comes to Intel's frequency optimized parts (see here: https://www.servethehome.com/second-gen ... -analysis/). It seems like AMD might have shaken up the stack a little with the launch of the EPYC 7371, causing Intel to drop the price for the 6242 and a re-positioning in the market for the 5217. I don't know what sorts of numbers these parts sell in, but I wonder if AMD was able to Intel a little harder in that segment than others to trigger such a large change.

*EDIT* I'll respond to your next post later tonight, but I have a long commute in front of me now.
 
Glorious
Gold subscriber
Gerbilus Supremus
Posts: 11763
Joined: Tue Aug 27, 2002 6:35 pm

Re: Big release day for Intel

Tue Apr 02, 2019 4:31 pm

Goty wrote:
If you can't get from point A to point B from that, I'm not sure how to proceed here.


I'm asking you to get me from point A to point B, and you failed. You said I was doing something, I asked how, and you just repeated the claim: Reading me the definition isn't describing how my conduct fits it.

I am attacking your position here, not your character. Even when I have discussed your motives, it was wholly pursuant to your own argument as explicitly stated: If you say you are addressing the lack of scrutiny of Intel's product vis-a-vis AMD's, yes, that makes "bias" a part of your argument. In fact, *YOU* were the one implicitly questioning everyone's else commitment to honesty and fair play irrespective of their actual arguments. (What, how dare we notice? :roll: )

Likewise, if your argument is ignorant of physical fact, it is not an irrelevant attack upon your person to simply state as such, especially when you rhetorically claim that no one cares.

Because if someone (non-trivially) does, as many in fact do, yes, that was an ignorant thing to say. Definitionally (as I see you are fond of them)

You, alone, bear responsibility for your irresponsible rhetorical flourishes. If you don't like where they lead you, the solution is stop taking that path, not to blame everyone you find on it afterwards.
 
Convert
Grand Gerbil Poohbah
Posts: 3451
Joined: Fri Nov 14, 2003 6:47 am

Re: Big release day for Intel

Tue Apr 02, 2019 5:51 pm

You really aren't benefiting the community by going to the extremes that you do.

The normal trolls are just your regular background noise found everywhere. Your reactions to trolls, or people just as highly opinionated as you, make you sound like a megaphone. Even when I agree with you, you make the whole opinion radioactive.
Tachyonic Karma: Future decisions traveling backwards in time to smite you now.
 
Waco
Gold subscriber
Grand Gerbil Poohbah
Posts: 3079
Joined: Tue Jan 20, 2009 4:14 pm
Location: Los Alamos, NM

Re: Big release day for Intel

Tue Apr 02, 2019 6:08 pm

chuckula wrote:
A very big launch even if the HBM equipped Icelake Xeons will make more headlines when they launch next year.

Got a roadmap for that?
Desktop: X570 Gaming X | 3900X | 32 GB | Alphacool Eisblock Radeon VII | Heatkiller R3 | Samsung 4K 40" | 1 TB NVME + 2 TB SATA + LSI (128x8) RAID
NAS: 1950X | Designare EX | 32 GB ECC | 7x8 TB RAIDZ2 | 8x2 TB RAID10 | FreeNAS | ZFS | LSI SAS
 
Glorious
Gold subscriber
Gerbilus Supremus
Posts: 11763
Joined: Tue Aug 27, 2002 6:35 pm

Re: Big release day for Intel

Tue Apr 02, 2019 6:32 pm

Convert wrote:
You really aren't benefiting the community by going to the extremes that you do.


While I can abstractly appreciate the wisdom of the concept that "Discretion is the better part of Valor", it clearly exceeds my practical grasp.

Alas.

Your point is graciously accepted.
 
NovusBogus
Silver subscriber
Graphmaster Gerbil
Posts: 1408
Joined: Sun Jan 06, 2013 12:37 am

Re: Big release day for Intel

Tue Apr 02, 2019 9:29 pm

I wonder if this means they're getting close to a successor to the Skylake based workstation Xeons. Those don't look so good next to Threadripper and the younger, core-boosted iterations of consumer silicon.
 
chuckula
Gold subscriber
Minister of Gerbil Affairs
Posts: 2106
Joined: Wed Jan 23, 2008 9:18 pm
Location: Probably where I don't belong.

Re: Big release day for Intel

Tue Apr 02, 2019 9:40 pm

Waco wrote:
chuckula wrote:
A very big launch even if the HBM equipped Icelake Xeons will make more headlines when they launch next year.

Got a roadmap for that?


This guy who bought 125 Petaflops [that's double precision petaflops, or as I call them: The awesome kind] worth of Knights Landing hardware does: https://www.hpcwire.com/2019/03/13/oil- ... 00-wafers/

That completely jibes with all the other rumors I've heard too, although I'll note that in those rumors nobody says that a single die solution will have 60 cores, although a 2-die connected solution can hit 60 - 70 cores easily.

And it makes sense. The only problem with a 56 core Cascade Lake is that 12 channels of RAM is frankly not enough in many situations. Augment that with HBM and an 8-channel DDR 5 platform becomes workable.
4770K @ 4.7 GHz; 32GB DDR3-2133; Officially RX-560... that's right AMD you shills!; 512GB 840 Pro (2x); Fractal Define XL-R2; NZXT Kraken-X60
--Many thanks to the TR Forum for advice in getting it built.
 
Waco
Gold subscriber
Grand Gerbil Poohbah
Posts: 3079
Joined: Tue Jan 20, 2009 4:14 pm
Location: Los Alamos, NM

Re: Big release day for Intel

Tue Apr 02, 2019 10:16 pm

chuckula wrote:
This guy who bought 125 Petaflops [that's double precision petaflops, or as I call them: The awesome kind] worth of Knights Landing hardware does: https://www.hpcwire.com/2019/03/13/oil- ... 00-wafers/

Forgive me for not trusting someone who claims US datacenters run on 120 volts as a reliable source.

Intel had a firesale on the KNL chips, I'm glad someone bought them up, but that doesn't make that person trusted by any means. He clearly didn't break any NDAs sharing that information (or he'd be in jail). Keep that in mind.
Desktop: X570 Gaming X | 3900X | 32 GB | Alphacool Eisblock Radeon VII | Heatkiller R3 | Samsung 4K 40" | 1 TB NVME + 2 TB SATA + LSI (128x8) RAID
NAS: 1950X | Designare EX | 32 GB ECC | 7x8 TB RAIDZ2 | 8x2 TB RAID10 | FreeNAS | ZFS | LSI SAS
 
dragontamer5788
Gerbil XP
Posts: 475
Joined: Mon May 06, 2013 8:39 am

Re: Big release day for Intel

Wed Apr 03, 2019 1:13 am

Waco wrote:
chuckula wrote:
This guy who bought 125 Petaflops [that's double precision petaflops, or as I call them: The awesome kind] worth of Knights Landing hardware does: https://www.hpcwire.com/2019/03/13/oil- ... 00-wafers/

Forgive me for not trusting someone who claims US datacenters run on 120 volts as a reliable source.


Sounds more like this guy is Australian and doesn't understand the US market. Seems like a believable mistake if all he's got is the internet as his reference point. I don't work at a datacenter, but L6-15 250V is what the server-room uses at my office (USA). The higher-efficiency of high-voltage would lead me to assume that most datacenter racks actually use 250V outlets. Offices typically use 120V (standard 5-15 American socket), but almost all computers will work with a 250V outlet so its not a big deal actually..

I still remember the first time I tried to plug a computer inside of the server-room by myself. I was very confused. The sysadmin was out on vacation and I was just... trying to make due without him at that time. Eventually he introduced me to the concepts of locking-plugs and such. Unfortunately, I had to purchase some equipment while he was gone... and ended up buying an (unnecessary) transformer. Ah well, n00b mistakes. The SysAdmin more or less laughed at me when he came back.

These are the sorts of things you only really learn when you walk into a room and try to plug a computer into the wall by yourself. For whatever reason, SysAdmins and electricians don't really talk about these details unless you ask them about it.
 
Topinio
Gerbil Jedi
Posts: 1748
Joined: Mon Jan 12, 2015 9:28 am
Location: London

Re: Big release day for Intel

Wed Apr 03, 2019 6:13 am

dragontamer5788 wrote:
I still remember the first time I tried to plug a computer inside of the server-room by myself. I was very confused. The sysadmin was out on vacation and I was just... trying to make due without him at that time. Eventually he introduced me to the concepts of locking-plugs and such. Unfortunately, I had to purchase some equipment while he was gone... and ended up buying an (unnecessary) transformer. Ah well, n00b mistakes. The SysAdmin more or less laughed at me when he came back.

I guess it was nice for him that you could do stuff, and that your "making do" proved he was worth having around and had expertise, that's got to be worth something to him.
Desktop: E3-1270 v5, X11SAT-F, 32GB, RX Vega 56, 500GB Crucial P1, 250GB Crucial MX500, 4TB 7E8, Xonar DGX, XL2730Z + G2420HDB
HTPC: i5-2500K, DH67GD, 6GB, RX 580, 250GB MX500, 1.5TB Barracuda, Xonar DX
Laptop: MacBook6,1
 
Glorious
Gold subscriber
Gerbilus Supremus
Posts: 11763
Joined: Tue Aug 27, 2002 6:35 pm

Re: Big release day for Intel

Wed Apr 03, 2019 7:12 am

dragontamer5788 wrote:
Sounds more like this guy is Australian and doesn't understand the US market. Seems like a believable mistake if all he's got is the internet as his reference point.


Right, and Waco is pointing out that if the internet is indeed his reference-point, he's not a trustworthy or useful source.

It's a totally believable mistake, as you say, but only for people completely unfamiliar with the subject. If you design datacenters, let alone work in just generic commercial construction, it just isn't. Saying that electricians don't want to "deal with" 240v in the USA is absurd--power is delivered to every residential building in the country as split-phase 240v. If you don't want to deal with every breaker-box in the country, it's time to consider other opportunities. :P

dragontamer5788 wrote:
I don't work at a datacenter, but L6-15 250V is what the server-room uses at my office (USA). The higher-efficiency of high-voltage would lead me to assume that most datacenter racks actually use 250V outlets. Offices typically use 120V (standard 5-15 American socket), but almost all computers will work with a 250V outlet so its not a big deal actually..


Right, nominal 240v is *very* common for non-datacenter server rooms in the USA. It's the same for me.

You can get even higher efficiencies if you design the entire facility for a custom purpose, like a datacenter can be, by using things like 208v 3 phase.

Google, almost two decades ago, starting taking things even further: Why have a centralized UPS? Why not just put a battery in the server? Why rectify individually in each server's PSU? Why not have rack-level DC distribution?

When you custom builld not just the building and the racks, but the servers themselves, the possibility space really increases.

dragontamer5788 wrote:
These are the sorts of things you only really learn when you walk into a room and try to plug a computer into the wall by yourself. For whatever reason, SysAdmins and electricians don't really talk about these details unless you ask them about it.


In the last 3 months we had to take two of our servers off-site for integration testing with a vendor system. The vendor site was just a generic office building complex (their neighbor was an ornery dentist).

Thus I had to stress that we had the right cables and they had the right outlets in the right places.

Like you said, sometimes such details get missed because they don't generally get talked about.

dragontamer5788 wrote:
but almost all computers will work with a 250V outlet so its not a big deal actually..


True, but some PSUs have a physical switch you'd want to flip if you actually did this. PSUs these days generally just seem auto-select, certainly the enthusiast ones people like us generally have.

Some rackmount servers PSUs will *only* work with 240v though, generally with a tolerance range of 20%. In such an event, you can't use 110 unless you get a different (they're typically modular, and for this very reason) PSU.

This was the case with the two servers I mentioned previously.



EDIT: fixed broken quote bracketing.
Last edited by Glorious on Wed Apr 03, 2019 9:13 am, edited 1 time in total.
 
Waco
Gold subscriber
Grand Gerbil Poohbah
Posts: 3079
Joined: Tue Jan 20, 2009 4:14 pm
Location: Los Alamos, NM

Re: Big release day for Intel

Wed Apr 03, 2019 8:58 am

Glorious wrote:
Right, and Waco is pointing out that if the internet is indeed his reference-point, he's not a trustworthy or useful source.

Exactly. Why would this guy who bought a lot of KNLs be any more of a valid source than any other Joe Blow on the Internet? Just buying a lot of CPUs and not understanding the market doesn't make you trustworthy (at least not in my mind).

I can't find anything on recent public roadmaps that says anything about HBM on Intel chips. Has anyone else seen otherwise? The only mentions of it in any form are for Phi, and that line is dead.


A little off topic, but we have 48V DC distribution on our smaller clusters, standard 208V for infrastructure, and 480V 3-phase for our big machines. The only 120v in any of the datacenters is at the walls for convenience of the cleaning crew.
Desktop: X570 Gaming X | 3900X | 32 GB | Alphacool Eisblock Radeon VII | Heatkiller R3 | Samsung 4K 40" | 1 TB NVME + 2 TB SATA + LSI (128x8) RAID
NAS: 1950X | Designare EX | 32 GB ECC | 7x8 TB RAIDZ2 | 8x2 TB RAID10 | FreeNAS | ZFS | LSI SAS
 
Glorious
Gold subscriber
Gerbilus Supremus
Posts: 11763
Joined: Tue Aug 27, 2002 6:35 pm

Re: Big release day for Intel

Wed Apr 03, 2019 9:36 am

Waco wrote:
Exactly. Why would this guy who bought a lot of KNLs be any more of a valid source than any other Joe Blow on the Internet? Just buying a lot of CPUs and not understanding the market doesn't make you trustworthy (at least not in my mind).

I can't find anything on recent public roadmaps that says anything about HBM on Intel chips. Has anyone else seen otherwise? The only mentions of it in any form are for Phi, and that line is dead.


Yeah, I think we're on safe ground completely discounting that article.

And I don't follow this really, but like you, I'm not aware of any such roadmap. And Intel is usually pretty good about that sort of thing.

Who is online

Users browsing this forum: No registered users and 2 guests
GZIP: On