Comment by wolf550e
12 hours ago
The amount of throughput you can get out of AMD EPYC zen5 servers for the price of a basic mainframe is insane. Even if IBM wins in single core perf using absurd amount of cache and absurd cooling solution, the total rack throughput is definitely won by "commodity" hardware.
These comments always come up with every mainframe post. It's not only about performance. If it were it would be x86 or pSystems (AIX/POWER). The reason customers buy mainframes is RAS (reliabililty, availability, scalability). Notice that performance is not part of RAS.
You and the parent are both "missing the point", which is sadly not talked about by the manufacturer either (IBM).
I used to work for Citrix, which is "software that turns Windows into a mainframe OS". Basically, you get remote thin terminals the same as you would with an IBM mainframe, but instead of showing you green text you get a Windows desktop.
Citrix used to sell this as a "cost saving" solution that inevitably would cost 2-3x the same as traditional desktops.
The real benefit for both IBM mainframes and Citrix is: latency.
You can't avoid the speed of light, but centralising data and compute into "one box" or as close as you can get it (one rack, one data centre, etc...) provides enormous benefits to most kinds of applications.
If you have some complex business workflow that needs to talk to dozens of tables in multiple logical databases, then having all of that unfold in a single mainframe will be faster than if it has to bounce around a network in a "modern" architecture.
In real enterprise environments (i.e.: not a FAANG) any traffic that has to traverse between servers will typically use 10 Gbps NICs at best (not 100 Gbps!), have no topology optimisation of any kind, and flow through at a minimum one load balancer, one firewall, one router, and multiple switches.
Within a mainframe you might have low double-digit microsecond latencies between processes or LPARs, across an enterprise network between services and independent servers its not unusual to get well over one millisecond -- one hundred times slower.
This is why mainframes are still king for many orgs: They're the ultimate solution for dealing with speed-of-light delays.
PS: I've seen multiple attempts to convert mainframe solutions to modern "racks of boxes" and it was hilarious to watch the architects be totally mystified as to why everything was running like slow treacle when on paper the total compute throughput was an order of magnitude higher than the original mainframe had. They neglected latency in their performance modelling, that's why!
The mainframe itself (or any other platform for that matter) is not magical with regards to latency. It's all about proper architecture for the workload. Mainframes do provide a nice environment for being able to push huge volumes of IO though.
6 replies →
> The real benefit for ... Citrix is: latency.
I understand your point about big iron, but where does something like Citrix reduce latency?
My estimation would be: Compared to a desktop Citrix adds distance and layers between the user and the compute/storage/etc., and the competition for resources on the Citrix server would tend to increase latency compared to the mostly idle desktop.
2 replies →
I’d love to read more about these projects. In particular, were they rewrites, or “rehosting”? What domain and what was the avg transaction count? Real-time or batch?
1 reply →
Maybe, but then you need to engineer the 99.99999% uptime yourself.
If it were actually cheaper, IBM wouldn’t be selling these machines so well.
They are mostly selling to the captive audience who is 40 years deep into COBOL and can't pull out until it falls on top of them.
Many of those machines are running Java workloads, also COBOL isn't the only mainframe language, and the best thing in terms of security is that they don't use C as systems language, rather saner stuff with proper arrays, strings and bounds checking.
That is why Unisys ClearPath MCP is still a thing, tracing back to its Burroughs 1961 heritage, security above all.
Well, you know what to do - just sell a system that runs COBOL on an x86 cluster or Kubernetes or whatever, and you'll make billions!
Be sure to allocate me a bunch of shares for giving you the idea.
I don't think they are actually selling those machines so well. They have a captive legacy customer base, who else is buying those?
From TFA: "Overall, Z is growing very healthily. LinuxONE is the fastest area of growth for us right now."
However, he didn't elaborate or give any examples. If I were the interviewer, I would have followed it with: "Oh?! Can you provide some examples for the readers who believe that you only sell to captive audiences?"
[dead]