Oh that's a weird way to do it; they used to have an x86 add on block for mainframes which was just a pile of x86 blades with some integration.

Z/OS for ARM then? ;-)

I’ve been running VM/370 and MVS on my RPi cluster for a long time now.

But I wonder if this is "much better" than x86 emulation or virt?

Is there really SW that's limited to (Linux) ARM and not x86?

Putting consumer grade (aka "commodity") hardware in a datacenter and running your infra on it is a bit of a meme, in the sense that it's not the only way of doing things. It was probably pioneered/popularized by Google but that's because writing great software was their "hammer", ie they framed every computing problem as a software problem. It was probably easier for them (= Jeff Dean) to take mediocre hardware and write a robust distributed system on top instead of the other way around.

There is, however, a completely different vision for how web infrastructure should be and that is to have extremely resilient hardware and simple software. That's what a mainframe is. You can write a simple and easy to maintain single process backend program, run it on a mainframe and be fairly confident that it can run without stopping for decades. Everything from the power supply to the CPU is redundant and can be hot swapped without booting the OS. Credit card transactions and banking software run on this model for example (just think about how insanely reliable credit card transactions are).

IBM has a monopoly in the second world. You could say the entire field of distributed systems is one big indie effort to break free of IBM's monopoly on computing.

A better question would probably what they don't do; just going off the wiki page (https://en.wikipedia.org/wiki/IBM) for recent history, they're in health care (imaging), weather, video streaming, cloud services, Red Hat, managed infrastructure (which branched off into a company called Kyndryl, which has 90.000 employees in 115 countries), warfare ("In June 2025, IBM was named by a UN expert report as one of several companies "central to Israel's surveillance apparatus and the ongoing Gaza destruction.""), etc etc etc.

Basically they do a lot, but they're not showy about it.

IBM has more revenue than Oracle even if we hear way less about it. 5 times smaller than Apple, thou. It also has more employees than Microsoft or Alphabet. But it has tighter profit margins than other tech companies.

IBM is not in consumer products nor services so we do not hear about it.

Early in my career I spent some years working at the biggest bank in Canada, they were (and still are) an enormous IBM customer. Hardware, software, consulting, and probably lots of other things I had no visibility into.

Beneath the countless layers of VMs and copious weird purpose built gear like Tandem and Base24 for the ATMs was a whole bunch of true blue z/OS powered IBM mainframes chugging through thousands and thousands of interlocking COBOL programs that do everything from moving files between partner banks all over the world, moving money between accounts, compounding interest, and extracting a metric shitton of every type of fee imaginable.

If you know z/OS there's work available until your retirement. Miserable, pointless, banal, and archaic legacy as fuck mainframe work.

https://en.wikipedia.org/wiki/Tandem_Computers

https://en.wikipedia.org/wiki/BASE24

https://en.wikipedia.org/wiki/Z/OS

Own Red-Hat, thus major contributions to Wayland, GNOME, GCC and Java, at very least.

Have their own Java implementation, with capabilities like AOT before OpenJDK got started on Leyden, or even Graal existed, for years had extensions for value types (nowadays dropped), and alongside Azul, cluster based JIT compiler that shares code across JVM instances.

IBM i and z/OS are still heavely deployed in many organisations, alongside Aix, and LinuxONE (Linux running on mainframes and micros).

Research in quantum computing, AI, design processes, one of the companies that does huge amounts of patents per year across various fields.

And yes a services company, that is actually a consortium of IBM owned companies many of each under a different brand (which is followed by "an IBM company").

I work for a big international corp. We pay IBM a blankest sum annually because it’s that hard to quantify just how much we rely on their services and licensing costs.

Licensing of course just being typical rent seeking behaviour but their services are valuable given the financial impact if one of their solutions goes down on us (which is very rarely)

Everything. They have done for decades, and will do for decades. And what IBM focus on is probably worth looking into.

IBM (imho) is in the absolute frontline in quantum computers. One could argue if the number of startups in QC means that there is an actual market or not. Companies that lives on VC or the valuation of their stock.

But IBM is not showy, not on the front pages, does not live on VC or stock valuation. IBM makes tons of money decade after decade from customers that are also not showy but makes tons of money. Banks, financial institutions, energy, logistics, health care etc etc. If IBM thinks these companies will benefit from using QC from IBM (and pay tons of money for it), there is quite probably some truth in QC becoming useful in the near future. Years rather than decades.

IBM have run the numbers and have decided that spending the money for engineering, research required is outweighs the money possible to earn on QC services. QCs powerful enough to run the QC-supported algorithms these companies need to make more tons of money. And it's probably not breaking RSA or ECC.

They make $8-9B a year (~90% profit margins) selling software to mainframes, which were deployed ages ago but still have to be maintained because critical COBOL business code was written on their systems - and migration is too riskly/costly.

To give you an idea:

- of the risk in regulated industries like banking: a UK bank was once fined *$62 million* for botching a mainframe migration and causing downtime. - of the difficulty and risk in non-tech industries: Australia once spent *$120 million* trying to migrate its social security system off mainframes... and failed.

Mainframes are not their only business, of course, but it's a major cash cow that's under appreciated. I, for one, didn't know that business keeps growing.

Coincidentally, I wrote about the topic of mainframes with relation to IBM's acquisition of Confluent here today: https://blog.2minutestreaming.com/p/ibm-confluent-acquisitio...

They design their own CPUs, and they sold $15b of hardware last year. Tellum ii in the z17 mainframe is a Samsung 5nm part.

What I don't get however is who'd use their custom accelerators for AI inference.

So they had $30 billion in software revenue last year and $15 billion in infrastructure against $20 billion in consulting.

You don't read much about IBM here, but this is the wrong site to look for them. A big chunk of IBM's business comes from other businesses outside the IT industry. You're more likely to read about IBM in the Wall Street Journal; Google finds "IBM" at wsj.com about 48000 times (it finds "oracle" there about 30000 times).

They design and build not one but two CPU architectures, s390/Z and POWER.

Both have been around for many years, but neither is obsolete, they're just not designed for consumer applications.

They still generate $10-15 billion per year in revenue.

I was surprised to find out they still have hardware repair technicians (extremely expensive but reliable: ~$400 per computer around 2022 iirc)

But yes they’re mostly enterprise/services/mainframes not anything overly consumer

I own their shares due to their Quantum Computing group

You can see their roadmap here:

https://www.ibm.com/roadmaps/

When you’re that large and established it’s very hard to die. I expect IBM to exist in some form pretty much forever

my company uses as400 and DB2 and pay for their servers. So they still make money from hardware too

Mainframes and consulting.

They own things like:

1. Red Hat Enterprise Linux, which is by far the most commonly deployed Linux variant among US Enterprise orgs.

2. Ansible

3. Podman

4. Hashicorp Terraform / Consul / Packer / Vagrant / Nomad / Etc.

5. Giant B2B services arm

6. Mainframe, which a lot of science organizations / governments / credit card companies still run. Sometimes you may have an IBM rep show up to replace a part on the mainframe you didn't even know was broken - very reliable, fault tolerant system.

7. The only service I know where you can rent Quantum computing time in the cloud

8. Probably a ton of other things I'm not even aware of.

9. Red Hat OpenShift - so if you're big enterprise running k8s on prem, there's a good chance it's OpenShift, especially in banking / finance / government.

They exist to swallow up profitable companies, extract any “unnecessary” overhead (like benefits, PTO, pay that isn’t rock bottom), and package into large enterprise licensing agreements.

I was shocked when IBM acquired Red Hat a few years ago. I had silently assumed at the time that Red Hat was far bigger than IBM nowadays, so the reverse would have made more sense to me.

They sell (managed) database appliances (on z and Power) and associated software (think the platform/HANA parts of SAP) - all state-of-the-art in the late 1990s but since then put on maintenance mode and it shows (a bit like oracle...). Their hardware is still cool custom built silicon and imo state of the art, but since k8s, high-speed-network and multi-TB-machines (for <100k$) are here and run Linux no new venture buys into that anymore (except for gulf states...). Before, when the competition was a cluster of Itanium/VMS or Sparc/Solaris and the associated contract, noone bought into that either at scale but also noone using IBM had a very compelling reason to switch everything around.

So essentially they sell new hardware and "support" to customers who have been in need to process tabular, multi-GB databases since when a PC was 128MB memory and have been doing electronic record-keeping since the 1970s. They also allow their ~hostages~, ehm, customers who trust them with their data to run processing near the data at a cost/in a cloud style billing model. That is so expensive though that every large IBM-shop has built an elaborate layer of JVMs, Unix and mirror-databases around their IBM appliances. Lately they bought Redhat and hashicorp and confluent thus taking a cut from the "support" of the abominiations of IT systems they helped birth for some more time to come (also remember the alternative JVM OpenJ9, do you all?).

I think the later a company started using centralized electronic record keeping, the higher the likelyhood they are not paying IBM anymore: commercial banks, governments and insurance started digitizing in the 60s (with custom software) and if the companies are old (or in US-friendly petrostates) they are all IBM customers. Corps using ERP or PLM offerings (so manufacturing and retail chains which are younger than banks) used to start digitizing a little later (Walmart only was founded in the 60s and electronic CAD started in the 80s) and while they likely used IBM in the past (SAP was big on DB2) they might not use it anymore (also it helps they usually bought the ERP or PLM from someone else). New Companies whose sole business was to run a digital-platform started on Unix (see Amazon who successfully fought to ditch Oracle even) or just built their whole platform (Google). If those companies predate Unix they usually fought hard to get rid of IBM (Microsoft, Amadeus)

Consulting/outsourcing services have been spun out to Kyndryl, so nowadays IBM only sells hardware, support for their products and ostensibly has some people left to develop their products... The days when that was a big thing and IBM produced all the stuff they sell support for now, have been long gone. A fun link to see how their "product development" operates nowadays is this discussion to bring gitlab-runners to z/OS: https://gitlab.com/gitlab-org/gitlab-runner/-/work_items/275... - tl;dr "hey you opensource company, we are IBM and managed to pay someone to port a go compiler to z/OS. Now we have a customer who wants to use gitlab with z/OS. Would you like to make your software part of our product offering?". A fun fact is that - even within IBM - access to the real mainframe seems to be very limited which shows a bit in the discussion linked above and also with an ex-Kyndryl-person saying: "oh, I once had a contract where we replaced the mainframe and we ran that on Linux-boxes inside IBM, because it was just cheaper that way. Just the big reporting was a bit slow, but the reliability was just fine"

>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.

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.

> 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

AI= Arm Ibm in that case

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.

You can run 1960s System/360 binaries unmodified on modern z/OS. The system also uses a lot of "high level assembler" and "system provided assembly macros" making a complete architecture switch extremely painful and complicated.

They called their new architecture "ESAME" for a while for a pretty obvious reason.

It's what Intel did with x86 a few decades before the modern smart phone.

Smartphones were a big boost, but they were already growing very rapidly before that.

PowerPC doesn't have the organic ecosystem that ARM has.

This certainly has been in the making for longer than the "everything we do must be for AI" bubble. In fact s390 has its own on-die inference engines and they have access to the same caching mechanisms as the main processor (which are quite insane).

Is modern ARM stuff done big-endian? because AIX is exclusively BE iirc

Is that such a silly notion?

No, it was Qualcomm who acquired Arduino. While they are an ARM licensee who make ARM chips, they are not ARM.