| ▲ | silvestrov 5 hours ago |
| > dual‑architecture hardware that helps enterprises run future AI and data intensive workloads with greater flexibility, reliability, and security I think we can ignore the "AI" word here as its presence is only because everything currently has to be AI. So why would IBM add ARM? > As enterprises scale AI and modernize their infrastructure, the breadth of the Arm software ecosystem is enabling these workloads to run across a broader range of environments I think it has become too expensive for IBM to develop their own CPU architecture and that ARM64 is starting to catch up in performance for a much lower price. So IBM wants to switch to ARM without making a too big fuzz about it. |
|
| ▲ | adrian_b 40 minutes ago | parent | next [-] |
| >So IBM wants to switch to ARM without making a too big fuzz about it. That was my first thought too, but it does not make sense, because if IBM would sell ARM-based servers nobody would buy from them instead of using cheaper alternatives. As revealed in another comment, at least for now their strategy is to provide some add-in cards for their mainframe systems, containing an ARM CPU which is used to execute VMs in which ARM-native programs are executed. So this is like decades ago, when if you had an Apple computer with a 6502 CPU you could also buy a Z80 CPU card for it, so you could also run CP/M programs on your Apple computer, not only programs written for Apple and 6502. Thus with this ARM accelerator, you will be able to run on IBM mainframes, in VMs, also Linux-on-ARM instances or Windows-on-ARM instances. Presumably they have customers who desire this. I assume that the IBM marketing arguments for this are that this not only saves the cost of an additional ARM-based server, but it also provides the reliability guarantees of IBM mainframes for the ARM-based applications. Taking into account that today buying an extra server with its own memory may cost a few times more than last summer, an add-in CPU card that shares memory with your existing mainframe might be extra enticing. |
|
| ▲ | rzerowan 4 hours ago | parent | prev | next [-] |
| Im thinking maybe as a compliment to x86 offerings and eventual displacement as a primary offering , i do not see them ditching POWER. The architecture might be non-standard and not very widespread however for what it does and workloads that are suited to it. I dont think any ARM design comes close , maybe Fujitsu's A64FX. |
| |
| ▲ | silvestrov 3 hours ago | parent [-] | | Marketingwise I think it is difficult for IBM to sell x86 systems as it is too easy for customers to compare performance to a standard Wintel server. Sun had the same problem after 2001 dotcom when standard PC servers became reliable enough to run web servers on. It's easier to sell "our special sauce" when building using a custom ARM platform. Then you have no easy comparison with standard servers. | | |
|
|
| ▲ | tempay 4 hours ago | parent | prev | next [-] |
| > ARM64 is starting to catch up in performance for a much lower price Why do you say "starting to"? arm64 has been competitive with ppc64le for a fairly long time at this point |
| |
| ▲ | adrian_b 28 minutes ago | parent | next [-] | | I do not think that I have seen any public benchmark for more than a decade that can compare ARM-based CPUs with IBM POWER CPUs. The recent generations of IBM POWER CPUs have not been designed for good single-thread performance but only for excellent multi-threaded performance. So I believe that an ARM CPU from a flagship smartphone should be much faster in single thread that any existing IBM POWER CPU. On the other hand, I do not know if there exists any ARM-based server CPU that can match the multi-threaded performance of the latest IBM POWER CPUs. At least for some workloads the performance of the ARM-based CPUs must be much lower, as the IBM CPUs have huge cache memories and very fast memory and I/O interfaces. The ARM-based server CPUs should win in performance per watt (due to using recent TSMC processes vs. older Samsung processes) and in performance per dollar, but not in absolute performance. | |
| ▲ | mbreese 2 hours ago | parent | prev [-] | | I thought PPC was supposed to be highly performant, but not very efficient. I didn’t think ARM (at least non-Apple ARM) was hitting that level of performance yet. I thought ARM was by far more efficient, but not quite there in terms of raw performance. But I could be wrong… I’m going from a historical perspective. I haven’t checked PPC benchmarks in quite a while. | | |
| ▲ | kjs3 28 minutes ago | parent [-] | | Are you guys sure you're not confusing product lines? PPC is a PowerISA architecture, but hasn't been pushing desktop/server level performance for, what, almost 20 years? It's an embedded chip now, and AFAIK IBM doesn't even make them any more. Power (currently "10th gen"(-ish)) is the performant aarchitecture, used in the computers formally known as i-Series, formerly known as RS/6000. It's pretty fast, not not price competitive. They aren't really the same thing. | | |
| ▲ | adrian_b 11 minutes ago | parent [-] | | "PowerPC" was a modification of the original IBM POWER ISA, which was made in cooperation by IBM, Motorola and Apple. Motorola made CPUs with this ISA. Apple used CPUs with this ISA, some made by IBM and some made by Motorola. While Motorola and Apple used the name "PowerPC", IBM continued to use the original name "POWER" for its server and workstation CPUs. Later IBM sold its division that made CPUs for embedded applications and for PCs, retaining only the server/workstation CPUs. However, nowadays, even if the official IBM name is "POWER", calling it "PowerPC" is not a serious mistake, because all the "PowerPC" ISA changes have been incorporated many years ago into the POWER ISA. So the current POWER ISA is an evolution of the PowerPC ISA, which was an evolution of the original 1990 POWER ISA. It is better to call it POWER, as saying "PowerPC" may imply a reference to an older version of the ISA, instead of referring to the current version, but the 2 names are the same thing. PowerPC was an attempt of rebranding, but then they returned to the original name. |
|
|
|
|
| ▲ | homarp 4 hours ago | parent | prev | next [-] |
| AI= Arm Ibm in that case |
| |
| ▲ | 3form 2 hours ago | parent [-] | | That's quite loaded already. They should consider calling it IBM ARM 64, IA-64 in short. | | |
|
|
| ▲ | formerly_proven 2 hours ago | parent | prev [-] |
| IBM has two architectures which are de-facto only used by them, s390x and ppc64le. They have poured a lot of resources into having open source software support those targets, and this announcement might mean they find it easier/cheaper going forward to virtualize ARM instead and maybe even migrate slowly to ARM. |
| |
| ▲ | mghackerlady an hour ago | parent | next [-] | | ppc64le has other machines. Raptor off the top of my head, but there's also that weird notebook project that seems to be talked about once every few years and probably won't ever happen and some pretty cool stuff in the amiga space (I don't know if that's strictly le but power is supposed to be bi-endian) | |
| ▲ | mbreese 2 hours ago | parent | prev [-] | | I think they see customers wanting to have the flexibility to move to ARM and this is the fastest way to say they support ARM workloads. Maybe this is a path for IBM to eventually use ARM chips down the road, but I see this as being more about meeting customers where they think the demand is today rather than an explicit guess for tomorrow. |
|
