News for hackers. Stuff that matters.
Re:Computing Performance on the Horizon by martinpw on Monday July 05, 2021 @01:56PM (#27740319)
Slide 26 is interesting - arguing that cloud providers have an advantage for future CPU design since they can analyze so many real world customer workloads directly.

In previous roles I have worked with CPU vendors who have been very keen on getting access to profiling data from our workloads for design optimization, and lamenting the fact that it was hard to get such data and they were often limited to synthetic benchmark workloads when tuning new designs.

So this argument does sound like a valid one, and does imply AWS etc will have significant advantages in future designs.

Re:Computing Performance on the Horizon by brendangregg on Monday July 05, 2021 @07:07PM (#27742677)
Thanks, I think it's the start of a new "cloud CPU" era.

Cloud vendors have already split workloads into different instance types, so directly analyzing their workloads and developing CPUs for each instance type will lead to further performance wins. This may prompt the creation of even more instance types just to further separate workload types for future CPU specialization.

In the future, products like AWS outposts may become far more desirable over commodity hardware, as customers know it provides access to specialized CPUs and their performance. It's a path for cloud computing vendors to own the datacenters as well.

(Note that my predictions are not based on any internal knowledge: I'm just describing what I personally would be doing if I were a cloud vendor.)

Re:Computing Performance on the Horizon by pm90 on Monday July 05, 2021 @10:31PM (#27743709)
In hindsight, I wonder why Intel or other chip makers didn’t invest more in datacenters/hosting companies just to get this valuable data.
Re:Computing Performance on the Horizon by brendangregg on Tuesday July 06, 2021 @01:29AM (#27744519)
Yes, this was in my predictions: Processor vendors may offer their own clouds (either built or acquired) to get the same level of workload access.
Re:Computing Performance on the Horizon by thechao on Monday July 05, 2021 @03:33PM (#27741159)
He sort of implies that “just better hardware” will peter out in the 2030s. I think he’s calling it at least 50 years too soon. Here’s why: (1) I think logic designers are still faffing about in term of optimizing their designs; and (2), I think there’s a lot of smart people thinking “incrementally” through what we’d consider paradigm shifts in HW implementation. That is, our fabs will just naturally segue into 3D, spintronics, etc. I think he even mentions 3D circuits? One thing a lot of people miss is that layout of the design is materially different in 3D vs 2D: in 2D layout is NP-hard (complete) without efficient polynomial approximations; in 3D layout is low-order polynomial. The reduction in layout complexity will allow us to design things that are unthinkable right now, due to layout constraints & wire congestion.
Re:Computing Performance on the Horizon by yjftsjthsd-h on Monday July 05, 2021 @04:41PM (#27741633)
> in 2D layout is NP-hard (complete) without efficient polynomial approximations; in 3D layout is low-order polynomial

Any chance you could explain to a novice why 3D is easier? To my naive intuition, it would have seemed like the more room to maneuver is offset by having more stuff to route.

Re:Computing Performance on the Horizon by blt on Monday July 05, 2021 @08:00PM (#27742971)
Intuitively, we feel that larger solution spaces should make a problem harder because there are more possible solutions to consider. And of course this is true if exhaustive search is the only algorithm.

But CS is full of problems where the smaller space is NP-hard and the larger one isn't. Integer linear programming is a prominent example.

To resolve the intuition, we can think of the larger space as an "unconstrained" problem where the solution space is somehow natural and nice for the problem. From this perspective, the smaller space looks like an added constraint. Adding constraints usually makes problems harder.

Re:Computing Performance on the Horizon by dragontamer on Monday July 05, 2021 @08:47PM (#27743238)
> Adding constraints usually makes problems harder.

Adding constraints can also make a problem easier. Which is why you've carefully worded your sentence in this manner :-)

You clearly know about those edge cases, but I felt it necessary to elaborate on this point: Additional constraints can make a problem easier OR harder, depending on the problem and the search space. Its very non-intuitive.

Re:Computing Performance on the Horizon by thechao on Monday July 05, 2021 @05:15PM (#27741878)
Try laying out a square with all the vertices connected in a plane: it’s not possible. You must move one of the wires “up” a layer. Which wire? Great layout minimizes layer transitions while also bunching together related HW blocks. The NP-completeness proof is related to work Knuth’s student (Plass?) did on laying out images in TeX.
Re:Computing Performance on the Horizon by hobofan on Monday July 05, 2021 @04:31PM (#27741551)
Aren't "big" 3D circuits unfeasible due to temperature limitations, though?
Re:Computing Performance on the Horizon by a1369209993 on Monday July 05, 2021 @04:52PM (#27741706)
> Aren't "big" 3D circuits unfeasible due to temperature limitations, though?

Less so than you'd think, since (as long as you can keep leakage current under control) heat is only generated when the circuits are active (barring leakage current, when bits are being flipped). So you can have arbitrarily large amounts of increasingly-rarely-used circuitry for various purposes.

The naive-but-easy-to-understand example would be having separate, optimal circuitry for each machine instruction - the total number of gates is O(N*M), but the number of gates activated on each clock cycle (and thus the amount of waste heat generated) is only O(N), so you can keep adding new, perfectly-hardware-accelerated instructions up to the limits of physical space. In practice it's more complicated and less of a nonissue, but it's not "big circuits are useful in proportion to their surface area, not their volume", it's more "big circuits are less useful than their volume alone would suggest". (You do hit physical limits like the Bekenstein bound[0] eventually, but that's far enough out that we mostly don't care yet.)

0: https://en.wikipedia.org/wiki/Bekenstein_bound

Re:Computing Performance on the Horizon by nradov on Monday July 05, 2021 @04:32PM (#27741560)
How are we going to cool those 3D chips?
Re:Computing Performance on the Horizon by thechao on Monday July 05, 2021 @05:19PM (#27741921)
The commenter above is correct: just stop toggling HW. We already do this to a great extent; we’re limited in the number of custom implementations because we can’t wire everything together. 3D chips will have a lot more “dark” logic than current chips, but will be orders-of-magnitude more efficient (& thus powerful) due to deep customization.

Also, remember the argument of my timeline is ~50–80 years out from now.

Re:Computing Performance on the Horizon by nradov on Monday July 05, 2021 @06:13PM (#27742297)
Dark logic isn't doing any useful work so what's the point? Sure you can include specialized circuitry for a bunch of rare cases, but that won't lift overall system performance much and will kill manufacturing yields.
Re:Computing Performance on the Horizon by thechao on Monday July 05, 2021 @07:05PM (#27742668)
The GPU in your computer consists of mostly-dark highly specialized circuits; it certainly seems to improve my computing experience.
Re:Computing Performance on the Horizon by ClumsyPilot on Monday July 05, 2021 @08:57PM (#27743287)
This is totally wishfull thinking. Customised logix means customised code.

The GPU has an h265 and h264 hardware video encoder that has no support on Mac and ia a bitch to get working on linux. Most software doesn't support it most software does not support GPU compute either, and we had that for a decade now.

Fuck, we had SIMD in every CPU for like 30 years, and out of 50 most popular programming languages how many even support it? 5?

Re:Computing Performance on the Horizon by thechao on Tuesday July 06, 2021 @07:25AM (#27746897)
Engineers will use the HW when it’s their only choice for year-over-year performance gains — I like to think of it as the “great unwinding”.
Re:Computing Performance on the Horizon by RantyDave on Monday July 05, 2021 @11:34PM (#27743997)
I would argue that most software doesn't need GPU compute.
Re:Computing Performance on the Horizon by AprilArcus on Monday July 05, 2021 @10:58PM (#27743837)
could it be fabbed around a lattice of thermally conductive elements, like a menger sponge?
Re:Computing Performance on the Horizon by justicezyx on Monday July 05, 2021 @06:58PM (#27742614)
It's already happening. Intel design now are significantly influenced by Google and Amazon's data center needs.
Re:Computing Performance on the Horizon by brendangregg on Monday July 05, 2021 @07:22PM (#27742768)
Sure, and I have regular meetings with processor vendors who want to understand our workload to better serve it. But there's no easy or fast way for them to get low-level data, including processor trace (cycle logs), across many customers. This can be vast amounts of data: Tbytes. If you are working on processors at a cloud vendor and want to know some low-level CPU detail, you could answer it immediately across a million customer workloads.
Re:Computing Performance on the Horizon by handrous on Monday July 05, 2021 @02:29PM (#27740623)
An interesting application of a now-familiar pattern: get lots of users, spy on them at massive scale, use those data to dominate some other market in a way that, at most, a single digit count of companies in the world could conceivably compete with (because none but they have anything like the data that you do). See also: everything to do with "AI".
Re:Computing Performance on the Horizon by uluyol on Monday July 05, 2021 @04:57PM (#27741749)
Amazon, Microsoft, and Google have massive applications and systems that they run, some of which they sell as a service. They have plenty of workload data without having to poke around user VMs.
Re:Computing Performance on the Horizon by handrous on Monday July 05, 2021 @05:17PM (#27741897)
You disagree with slide 26, then?