Live data from Hacker News

Intel Confronts Potential ‘PR Nightmare’ With Reported Chip Flaw

bloomberg.com

551–560 of 567 posts

Re: Intel Confronts Potential ‘PR Nightmare’ With Reported Chip Flaw

#551
post #31
post #23

Earlier quoted context omitted.

Unless demand is fully inelastic, if you up the price fewer people will buy. As such, this should hit the cloud provider profits.

> Unless demand is fully inelastic It's mostly inelastic. People don't buy resources unless they need to do work. Alternatives, such as on-prem hardware, are affected as well. The only work that would be affected would be the work that becomes unprofitable at a 10-30% hardware cost increase and this is probably marginal enough to be ignored.

> It's mostly inelastic. People don't buy resources unless they need to do work.

I work on a system (at Google) with significant hardware cost. Fundamentally, the work my system is doing needs to be done. But the time we spend improving efficiency is hugely elastic. I look at a CPU profile or request flow, have an idea to improve it, look at how much machine resources we're spending, guesstimate how long it will take us to implement and maintain, and use a chart to see if my idea is worth my team's time or not.[1]

If the resources get more expensive or simply impossible to acquire,[2] we'll optimize more.

[1] There are other considerations (will it make the code more complex, thus increasing risk of a security/privacy flaw? what about opportunity cost given that it's hard to hire more engineers and scale a team?) but that's a reasonable mental model.

[2] There's a global RAM and SSD manufacturing crunch already, and if lots of CPUs are replaced due to these vulnerabilities or simply are no longer enough, that's gonna be a big crunch as well. If you're one of the few biggest cloud providers, I think you can't just replace / add tons more hardware than planned without dramatically increasing the price per unit for everyone.

Re: Intel Confronts Potential ‘PR Nightmare’ With Reported Chip Flaw

#552
post #482
post #388

Earlier quoted context omitted.

> The hit I see is by providers of 'serverless' computing, since they charge per request and have their margins reduced. AWS, Azure, and GCP all bill serverless with a combination of per-request fees and compute (GB-seconds), so I'd expect the entire hit to be passed on to the user since this will cause increased compute time for each request. N requests that used to average 300ms each will now be N requests that ave…

I don't understand what exactly you're saying. All of those services have serverless services, but they also have server based instances which abstract compute to amount of cores and RAM rather than CPU cycles. And most use is out of the services which aren't serverless.

Then it's because you don't know how modern cloud works.

Re: Intel Confronts Potential ‘PR Nightmare’ With Reported Chip Flaw

#554
post #31

Earlier quoted context omitted.

> Unless demand is fully inelastic It's mostly inelastic. People don't buy resources unless they need to do work. Alternatives, such as on-prem hardware, are affected as well. The only work that would be affected would be the work that becomes unprofitable at a 10-30% hardware cost increase and this is probably marginal enough to be ignored.

> It's mostly inelastic. People don't buy resources unless they need to do work. I work on a system (at Google) with significant hardware cost. Fundamentally, the work my system is doing needs to be done. But the time we spend improving efficiency is hugely elastic. I look at a CPU profile or request flow, have an idea to improve it, look at how much machine resources we're spending, guesstimate how long it will take…

> But the time we spend improving efficiency is hugely elastic.

In the context of the above posts, "elastic" is an economic term used to refer to the demand curve. There is no supply/demand curve when it comes to internal technical decisions for optimization =)

I would say your work is highly correlated to the price of hardware per some unit of performance AND the amount of work that you need to complete AND the amount of hardware already available AND the cost of labor needed for optimization.

I imagine in your case, since these systems are tightly controlled, you can probably run unpatched without taking a substantial risk.

Re: Intel Confronts Potential ‘PR Nightmare’ With Reported Chip Flaw

#555
post #552
post #482

Earlier quoted context omitted.

I don't understand what exactly you're saying. All of those services have serverless services, but they also have server based instances which abstract compute to amount of cores and RAM rather than CPU cycles. And most use is out of the services which aren't serverless.

Then it's because you don't know how modern cloud works.

I now see that my misunderstanding was about who exactly the users were and who the provider was. My opinion of provider was only GCE, AWS, etc. while the commenter I believe when talking about providers included users of those services (who again were providers of serverless services).

Re: Intel Confronts Potential ‘PR Nightmare’ With Reported Chip Flaw

#556

Earlier quoted context omitted.

I see - so this is more accessible? We're all fucked, then?

The "fear" with ME is more that governments or someone like that could spy on us... (not really a "threat" to most people in Western Countries). But still better to get rid of it. Not an imminent threat per se. This bug is a true security bug that has been lurking around for 10 years... and it could have been already abused. This is double scary because this attack does not leave any traces behind and it allows any a…

OS updates will fix this - but the bug remains on the chip, right? Hm, I guess I'll check for updates =)

Re: Intel Confronts Potential ‘PR Nightmare’ With Reported Chip Flaw

#557

Earlier quoted context omitted.

It's a good thing CPU is fairly compressible. Unless you meter it very carefully, you'll see the performance hit and it'll not impact you that much. Very few of my physical boxes are over 70% CPU utilization on a daily average. It's, however, really bad if you sell CPU cycles for a living. You just lost between 5 and 30% of your capacity. If you have a large building, you just lost part of your parking lot to the Int…

About 10 years ago I was mentored by a guy who was an utter wizard at queuing theory, and who bugfixed a whole bunch of nasty issues in cellular telecoms hardware through his understanding of how queuing theory impacted code execution. TL:DR - queue behaviour gets nonlinear as you approach the theoretical max load. If you are running your processors at a high load, even a small change in code throughput makes a huge…

I believe I saw a strangeloop talk about this specific issue in Clojure. The talk giver was talking about channels, not queues though.

Re: Intel Confronts Potential ‘PR Nightmare’ With Reported Chip Flaw

#558
post #398

Earlier quoted context omitted.

According to this patch comment, AMD x86 chips are not affected: https://lkml.org/lkml/2017/12/27/2

Sure but who is using AMD chips in place of Intel server chips? If company A competes in the widget market against company B and they both built their server infrastructure on Intel then neither company gained an advantage due to a performance degradation in Intel hardware.

> Sure but who is using AMD chips in place of Intel server chips?

Well... Everyone who bought AMD. Some people managed to see beyond the hype and go for the optoon that made sense.

Re: Intel Confronts Potential ‘PR Nightmare’ With Reported Chip Flaw

#559
post #398

Earlier quoted context omitted.

Sure but who is using AMD chips in place of Intel server chips? If company A competes in the widget market against company B and they both built their server infrastructure on Intel then neither company gained an advantage due to a performance degradation in Intel hardware.

> Sure but who is using AMD chips in place of Intel server chips? Well... Everyone who bought AMD. Some people managed to see beyond the hype and go for the optoon that made sense.

The overwhelming majority of the cloud runs on Intel. Saying AMD is slightly better off doesn't really help if my systems are built on Intel. This is the case for most people.

What hype are you referring to? Are you suggesting the people who bought AMD knew this was a problem for Intel?

Re: Intel Confronts Potential ‘PR Nightmare’ With Reported Chip Flaw

#560

Earlier quoted context omitted.

You're doing it wrong if you're day trading stocks... dollar cost averaging is your friend when it comes to volatility and well if you're investing you're doing so because of long term belief in the company so it should be easy to hold and not worry about it.

"dollar cost averaging is your friend" Research suggests otherwise: https://personal.vanguard.com/pdf/ISGDCA.pdf Diversification is a more effective strategy for dealing with volatility.

I only read page one of that link (thanks for sharing), but it appears to limit its scope to a specific scenario: given a lump sump of money (a windfall), it is it better to invest all at once or dollar cost average over a period of time? It is not recommending against dollar cost averaging in general; it is recommending against it for windfalls. Diversification is also not mentioned in the article.
Post reply on HN