But purely from performance point of view, as in throughput, and also quality of service, the case is clear for the mainframe.
No, it isn't clear. Show the benchmarks.
No, it isn't clear. Show the benchmarks.
We can start here:
"IBM z15 Performance of Cryptographic Operations"
https://www.ibm.com/downloads/cas/6K2653EJ
Can your x86 do this?
Edit: "Introducing the new IBM z15 T02"
https://www.ibm.com/training/pdfs/The_New_IBM_z15_A-technica...
==========
5.2 GHz core frequency
• On-chip compression accelerator (NXU)
• On Core L1/L2 Cache
• L2-I from 2MB to 4MB per core
• On chip L3 Cache
• L3 from 128MB to 256MB per chip
960 MB shared eDRAM L4 Cache
=========
https://github.com/IBM/IBM-Z-zOS/blob/main/zOS-Education/Hin...
https://community.ibm.com/HigherLogic/System/DownloadDocumen...
So, I am guessing actual head to head benchmarks must be quite bad.
Your latest Intel® Xeon® E-2386G Processor, not available yet...has a base frequency of 3.50 GHz and a Max Turbo Frequency 5.10 GHz
So even with the latest...your Turbo frequency... is the base from the mainframe core...
And if you happen to be running a memory bound benchmark, with the massive caches in the mainframe, you will be crushed.
I can easily write a memory bound benchmark that makes all CPU cache irrelevant. The only thing that actually matters is real world performance which is why actual benchmarks are so critical.
> Xeon® E-2386G Processor
That’s a chip from 2018. Anyway, you can talk about CPU clock speeds all day, but it doesn’t mean much between different CPU families.
Wrong choice indeed the Xeon® E-2386G Processor its not what I was looking for. Instead of going for the latest lets review the latest offers from Intel:
https://en.wikipedia.org/wiki/List_of_Intel_processors
A few models that have a peak turbo frequency approaching the sustained frequency of Z core. Also, of course due to different architectures, internal caches, we should not be comparing purely on frequency. Although the mainframe is again wining here in frequency and cache sizes...
Also agree that what counts are real workloads with real commercial applications. I have so far posted at least three but here is another one:
"IBM MQ for z/OS on z15 Performance"