On 19/04/2018 12:00 PM, Paul Gilmartin wrote:
On Thu, 19 Apr 2018 10:21:20 +0800, David Crayford wrote:
On 19/04/2018 5:19 AM, Cheryl Watson wrote:
The z13 and z14 can run Java faster than any other processor.
Do you mean any other mainframe processor? My observations are that Java
runs significantly faster on a x86 Linux system compared to our z13s
running z/OS with a zIIP.
gcc? Has anyone a data point for similar compilers on Linux for zseries and
Linux for x86?
Is the C compiler on z/OS zIIP-eligible?
I've yet to see a Java workload run faster on z/OS then on x86. And our
x86 servers are heavily virtualized using Hyper-V. Our zIIP runs at
below 10% so there is plenty
of capacity available. It would be interesting to compare the
performance of Java running on z/OS vs zLinux.
It's not just Java worksloads. I ran a large C++ compile that took over
an hour on z/OS which ran under two minutes on x86/Linux. Our z13s
rarely runs at over 50%.
I've seen similar results for GNU configure. I attribute this to both compiler
and general process fork() overhead.
Are more address space creations nowadays due to UNIX fork() or batch
initiators? (Of course this is site-dependent.) I suspect AS creation is
still optimized for batch job steps ("BPXAS ON INTRDR"). This may be
suboptimal.
The compiles take forever. I recently had a meeting with the compiler
guys from Toronto and I asked them if a cross-compiler was on the radar.
That didn't like that question!
I would suggest the performance of the compiler is down to the software
stack it runs on.
-- gil
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN