41–50 of 135 posts
Earlier quoted context omitted.
They were pretty clear at Hot Chips that it's native
"code translation in hardware" microcode is called "native"
Earlier quoted context omitted.
At this point no one is buying into mainframes are they? So they are just selling to the existing pool of users for which the price just has to come in under the cost/risk of rewriting all their ancient stuff.
Someone has to be buying mainframes. Otherwise, there's no point in continuing to develop them. Shareholders would revolt.
Huh. ARM instead of ppc64le?
ARM has better software support for AI applications and half of their presentation was about their inference accelerators that can go in the mainframes (and POWER machines). IBM mainframes are almost designed by their users. The previous generation skipped a lot of speed boost on the CPU side because their users didn’t want the machine to blow over their power delivery limits. Now, with their architecture behind it,…
Earlier quoted context omitted.
To be fair, IBM did this in the 90s with the PowerPC 615, which had PowerPC and x86 decoders on the same core.
But those ARM and s390x decoders have a “Telum III” architecture behind them with a cache that’s pretty innovative. IBM has never stopped innovating. It’s just that most people can’t afford their machines.
That being said I think it's a natural consequence of the difference between a mainframe and commodity servers. A mainframe is going to be running pretty disparate workloads simultaneously, so it makes sense to steal from your neighbor if they aren't using their cache. Whereas it's more likely that a commodity server is just running the same server on each core, and if you have a different workload, you pick a different shape of server to run it on. There are pros and cons to both.
I do wonder about the spectre consequences of borrowing cache lines from other cores though.
This feels like a baby step towards Arm being able to emulate z/Arch workloads, maybe with a bit of secret sauce for certain specific operations, which doesn't seem very much like IBM.
I thought about that, too. I think they’re doing this for the same reason IBM has supported Linux LPARs: since a lot of customers who lease System/z currently probably get overprovisioned hardware that they’ll never fully use, why not try a last-ditch attempt to sell the excess capacity as ARM LPARs? Now you’re thinking like a Big Blue sales rep.
Earlier quoted context omitted.
ARM has better software support for AI applications and half of their presentation was about their inference accelerators that can go in the mainframes (and POWER machines). IBM mainframes are almost designed by their users. The previous generation skipped a lot of speed boost on the CPU side because their users didn’t want the machine to blow over their power delivery limits. Now, with their architecture behind it,…
At least back in the day, there was talk about Nvidia + ppc64le. Wasn't there even a supercomputer with that setup? But I guess that has fallen by the wayside.
Earlier quoted context omitted.
More like mid-2000s, even with the least charitable interpretation. There was a time when IBM dominated console CPUs for a massively successful generation (PS3, Xbox 360, Nintendo Wii).
It's a little more complex than that, IMO. For starters, IBM did also do the CPU for the GameCube, and that was likely a part of the Wii using an upgraded PPC arch for that. On the flipside, there is the theory (I think even Copetti brings it up in their XBox 360 Architecture breakdown [0]) that IBM using the Cell PPE for the 360's tri-cores left a sour taste in Toshiba, but more-so Sony's mouths. I think the big 'X…
Earlier quoted context omitted.
Wow, you’re right. Xeon vs Xenon I had always assumed with chip names that similar it was an Intel product. How on earth didn’t they get sued?
I'm guessing that they were safe in that it was more or less a 'codename' for a specific product's CPU that nobody else could buy, versus a product out on the market, left enough clear separation to avoid Trademark complaints. The combined might of IBM and Microsoft's legal teams, even (possibly even moreso?) then, was better than Intel's to be sure.
Earlier quoted context omitted.
"code translation in hardware" microcode is called "native"
Yep. All modern Intel and AMD architectures, save maybe some Atom variants, effectively JIT compile the incoming instructions. It's part of the reason the faster-than-static Java/JVM promises never came to fruition. The goal posts moved after the mainstream chips coopted the approach, in a sense, and they continue to move as pipelines get deeper and broader and transistor count goes up. Transmeta was a classic failur…
Intel and AMD were more or less 'Jitting' since the PPro and K5, and that's ignoring NexGen's Nx586 (which was shipping in hardware about a year before the PPro).
All of them are way before Transmeta shipped a CPU. Where Transmeta tried to innovate was by using VLIW rather than a RISC-like core, along with their other special tech (Which they later licensed to lots of companies, including Intel,) to provide reasonable mobile performance with a low power draw. Oh, there's also the bit where Transmeta CMS is much more software based (likely, partially to push enough to software and avoid an x86 CPU license lawsuit) but even modern Intel chips AFAIK are still doing more translation on the CPU layer.
Earlier quoted context omitted.
But those ARM and s390x decoders have a “Telum III” architecture behind them with a cache that’s pretty innovative. IBM has never stopped innovating. It’s just that most people can’t afford their machines.
Yeah, their cache work is undeniably really cool. That being said I think it's a natural consequence of the difference between a mainframe and commodity servers. A mainframe is going to be running pretty disparate workloads simultaneously, so it makes sense to steal from your neighbor if they aren't using their cache. Whereas it's more likely that a commodity server is just running the same server on each core, and i…