Live data from Hacker News

Intel's make-or-break 18A process node debuts for data center with 288-core Xeon

tomshardware.com

231–240 of 303 posts

Re: Intel's make-or-break 18A process node debuts for data center with 288-core Xeon

#231

More Intel vaporware. Seriously, their other 18A product, panther lake, supposedly "launched" January 18th. It's been 1.5 months and I still can't go and buy any panther lake laptop except from dell.com. Why are they like this? I'll believe it when I see it. Also about "make-or-break": they've been saying this for all of Intel's products since at least 2022 *yawn*

AFAIKS there are 18A laptops from Lenovo and HP for the first two I checked. Or do you mean if you click through and buy one it gets put on some indeterminate backorder?

Re: Intel's make-or-break 18A process node debuts for data center with 288-core Xeon

#232
post #60
post #27

With packages like this (lots of cores, multi-chip packaging, lots of memory channels), the architecture is increasingly a small cluster on a package rather than a monolithic CPU. I wonder whether the next bottleneck becomes software scheduling rather than silicon - OS/runtimes weren’t really designed with hundreds of cores and complex interconnect topologies in mind.

There definitely are bottlenecks. The one I always think of is the kernel's networking stack. There's no sense in using the kernel TCP stack when you have hundreds of independent workloads. That doesn't make any more sense than it would have made 20 years ago to have an external TCP appliance at the top of your rack. Userspace protocol stacks win.

> Userspace protocol stacks win.

No they don't. They are horribly wasteful and inefficient compared to kernel TCP. Also they are useless because they sit on top of a kernel network interface anyways.

Unless you're doing specific tricks to minimize latency (HFT, I guess?) then there is no point.

Re: Intel's make-or-break 18A process node debuts for data center with 288-core Xeon

#233
post #24

So TLDR is it competitive? What are the dimensions and dynamics here vs EPYC?

Not competitive at all. It's easily visible on the laptop lines, where the same GPU manufactured on TSMC has 3 times the power/performance ratio compared to the Intel one. Putting more cores is just another desperate move to play the benchmark. Power is roughly quadratic with frequency, every time you fall behind competition, you can double the number of cores and reduce the frequency by 1.414 to compensate. Repeat a…

That's very interesting if it's the same GPU and perf/W is that much worse. Where are these numbers published please?

Re: Intel's make-or-break 18A process node debuts for data center with 288-core Xeon

#234

More Intel vaporware. Seriously, their other 18A product, panther lake, supposedly "launched" January 18th. It's been 1.5 months and I still can't go and buy any panther lake laptop except from dell.com. Why are they like this? I'll believe it when I see it. Also about "make-or-break": they've been saying this for all of Intel's products since at least 2022 *yawn*

AFAIKS there are 18A laptops from Lenovo and HP for the first two I checked. Or do you mean if you click through and buy one it gets put on some indeterminate backorder?

link?

Re: Intel's make-or-break 18A process node debuts for data center with 288-core Xeon

#235

These sorts of core-density increases are how I win cloud debates in an org. * Identify the workloads that haven't scaled in a year. Your ERPs, your HRIS, your dev/stage/test environments, DBs, Microsoft estate, core infrastructure, etc. (EDIT, from zbentley: also identify any cross-system processing where data will transfer from the cloud back to your private estate to be excluded, so you don't get murdered with egr…

on prem = capex

cloud = opex

The accounting dept will always win this debate.

Re: Intel's make-or-break 18A process node debuts for data center with 288-core Xeon

#236

Earlier quoted context omitted.

AFAIKS there are 18A laptops from Lenovo and HP for the first two I checked. Or do you mean if you click through and buy one it gets put on some indeterminate backorder?

link?

https://www.lenovo.com/us/en/p/laptops/yoga/yoga-slim-series...

https://www.lenovo.com/us/en/p/laptops/ideapad/ideapad-slim-...

EDIT: Changed the lenovo link to a non-preorder laptop (was https://www.hp.com/us-en/shop/pdp/hp-omnibook-ultra-laptop-n...)

Re: Intel's make-or-break 18A process node debuts for data center with 288-core Xeon

#237

Earlier quoted context omitted.

And you think just anyone can set that up? No sys admin/infra guy needed? Seems pretty risky.

I mean not just anyone , but its far less complicated than dealing with arcane iptables commands. And yet far more powerful, being able to just say "instances like this can talk to instances like this in these particular ways, reject everything else". Don't need subnet rules or whatever, its all about identity of the actual things. Meanwhile lots of enterprise firewalls barely even have a concept of "zones". Its prac…

> I guess with on-prem stuff things are often less ephemeral, so there's slightly less need

Kubernetes is running on bare metal quite a lot of places.

Re: Intel's make-or-break 18A process node debuts for data center with 288-core Xeon

#238

Earlier quoted context omitted.

link?

https://www.lenovo.com/us/en/p/laptops/yoga/yoga-slim-series... https://www.lenovo.com/us/en/p/laptops/ideapad/ideapad-slim-... EDIT: Changed the lenovo link to a non-preorder laptop (was https://www.hp.com/us-en/shop/pdp/hp-omnibook-ultra-laptop-n... )

“Preorder”

Re: Intel's make-or-break 18A process node debuts for data center with 288-core Xeon

#239

These sorts of core-density increases are how I win cloud debates in an org. * Identify the workloads that haven't scaled in a year. Your ERPs, your HRIS, your dev/stage/test environments, DBs, Microsoft estate, core infrastructure, etc. (EDIT, from zbentley: also identify any cross-system processing where data will transfer from the cloud back to your private estate to be excluded, so you don't get murdered with egr…

The main cost with on-prem is not the price of the gear but the price of acquiring talent to manage the gear. Most companies simply don't have the skillset internally to properly manage these servers, or even the internal talent to know whether they are hiring a good infrastructure engineer or not during the interview process. For those that do, your scaling example works against you. If today you can merge three ser…

While I agree with you, some solutions, such as Oxide Computing could come pretty close to having all the ease of cloud, one whole rack of computers at a time.

Re: Intel's make-or-break 18A process node debuts for data center with 288-core Xeon

#240

Earlier quoted context omitted.

https://www.lenovo.com/us/en/p/laptops/yoga/yoga-slim-series... https://www.lenovo.com/us/en/p/laptops/ideapad/ideapad-slim-... EDIT: Changed the lenovo link to a non-preorder laptop (was https://www.hp.com/us-en/shop/pdp/hp-omnibook-ultra-laptop-n... )

“Preorder”

Oh, there are so many lenovo ones I just picked one at random and didn't notice it said preorder. The HP link doesn't say that though, right? Here's another lenovo one that doesn't say preorder.

https://www.lenovo.com/us/en/p/laptops/ideapad/ideapad-slim-...

Post reply on HN