Hacker News Reader

Comment by dheera | Hacker News Reader

I hate these hybrid configurations, I recently rebuilt my desktop and jumped from Intel to AMD (7950X, not 7950X3D) specificially because I don't want a hybrid desktop CPU. All of Intel's 14th gen desktop offerings are hybrid. I hope AMD doesn't get on that bandwagon.

I do a lot of highly parallelized image processing and preprocessing for machine learning and I don't want any cores holding other cores back.

View on HN

As your workload grows in thread count, processors become limited by memory bandwidth per socket, as well as power and thermal constraints. The 'big' cores can only perform at their peak frequency, power draw and IPC when not in a high multicore workload. So you end up wasting silicon resources, power in an all-big core design. Instead if you make the last several cores to be turned on smaller and optimized for a lower power state, you can fit more of them in, and achieve higher overall thread throughput for a given die size or power/thermal envelope.

So imagine you set a given budget for silicon die size and chip power. You might then have 3 design possibilities. Say 8 big cores, 16 little or 12 in hybrid. The hybrid design should be able to match the all-big in single/lightly threaded loads, while beating the all-big in highly multithreaded. These tradeoffs of course work best in form-factors that are constrained by cost or battery power. Which is where you see amd using them in consumer. Or by space/power where you see AMD giving choice of small/dense core options in Epyc.

Do you have a URL or something that reviews this in greater detail? Perhaps in a laptop scenario?

And what happens if you run something that's embarrassingly parallel and you need all the juice you can get from your cores. With hybrids you have to rely on the OS to decide when it will utilize everything and when not which in some heavy loads could cost you performance wise.

The OS doesn't decide how many threads the application spawns. Is there a hybrid system you're aware of where the OS will refuse to schedule N threads across N cores due to the cores not all being the same type?

The only such example I'm aware of is Intel's recently-launched Meteor Lake laptop chips that include an extra two low-power E cores on a separate chiplet from the rest of the CPU cores, running at half the performance of the regular E cores and with no L3 cache, so whether they're used or not only makes a few percent difference to embarrassingly parallel workloads.

I didn’t say that the OS refuses to utilize cores. But I’ve seen examples of embarrassingly parallel tasks that utilize 80% of the cpu to run faster on CPUs with only performance cores when compares to hybrids with the same number of threads. There might be some other reason for that, but my gut feeling is to stick with all performance designs.

> I’ve seen examples of embarrassingly parallel tasks that utilize 80% of the cpu to run faster on CPUs with only performance cores when compares to hybrids with the same number of threads

The "same number of threads" bit is throwing me off. A hybrid config will generally have more CPU cores allowing it to handle more threads for a chip of the same size, so testing with a fixed thread count can only tell you that the performance of each small core is less than the performance of a large core. But maybe the examples you have in mind are complicated by SMT/HyperThreading (which Intel has on P cores but not E cores, while AMD has it on both)?

I only mention threads because it's the only way to draw comparisons between dissimilar cpus.

You already specified that we're discussing embarrassingly parallel tasks, so putting a constraint on thread count is of no use here. We can compare on the basis of performance, power and die size.

The smaller cores are not really much different from having other applications / kernel / device drivers running concurrently with your workload.

Bad design: see N cores, split work into N partitions, start N threads, each thread processes one partition.

Good design: see N cores, start ~N threads, each thread takes 1/M of work at a time, M>>N.

> The 'big' cores can only perform at their peak frequency, power draw and IPC when not in a high multicore workload.

This is really not true. With a decent motherboard and sufficient cooling, you can force the big cores to stay virtually locked at peak clock speeds. Der8auer does it for a Ryzen 8000 APU here just as an example: https://youtu.be/VNYx72Elgss?t=335

Putting a laptop chip into a desktop platform and giving it water cooling and unlimited power delivery isn't a particularly strong counterexample. That's nowhere near the operating point the silicon was designed for, and the fact that AMD shipped a few SKUs that enable such usage as an afterthought doesn't disprove much.

It is hybrid but the only difference is that the little cores are clocked lower.

It is pretty common for a CPU to have different clock speeds depending on how many cores are active (not to mention AVX clocks, so clock speed is just a thing that bounces around). This doesn’t seem drastically different.

I concur with this and did the same. I think the schedulers are still not mature enough to properly handle these slower efficiency cores, and then you wind up getting frustrated when one of your tasks gets offloaded to one or more of these unexpectedly.

Minor quibble:

I have an Intel N100-based mini PC. As far as I can tell the N100 is not hybrid... because it only uses E-cores.

Have you actually tested them, though? It seems to me like having more threads with lower clocks would be an excellent compromise for your use case.

Yes, if they are all identical cores, and this is where GPUs shine.

But if you have two different kinds of cores with different clocks and no control over the scheduler it is not a good compromise.

The interesting thing is that even on highly dense workloads, the benchmarks and real world tests are showing that these hybrid core architectures are doing surprisingly great things.

I have more reading to do on the AMD side, so if you have more info than I take this with a grain of salt, but at least on the Intel side of the camp, the performance of a fully saturated E core is only slightly worse than its P core buddy, with 2x E cores meeting or exceeding the work output of a single P core + its hyperthread.

Even if the only purpose for the E cores is to run OS and other background workloads while your big processing job is getting unfettered access to the P core array, I imagine you'd see pretty great gains here.

Reply on news.ycombinator.com