On 2023-04-27, Single Stage to Orbit <alex.buell@munted.eu> wrote:
> On Thu, 2023-04-27 at 17:31 +0000, Simon Clubley wrote:
>> If those are the kernel figures for an Intel box, I would love to see
>> what they are for an AMD box...
>
> Point me in the direction of a copy of VUPS.COM and I can run these
> benchmarks.

VUPS.COM is interesting, but it isn't the thing that's showing up
the extra kernel mode overheads for real-world activities.

Mark's numbers shown here:

https://wasd.vsm.com.au/info-WASD/2023/0077

with real-world I/O activity, are showing major increases in the percentage
of kernel mode overhead time consumed.

I would be interested in seeing what VSI have to say about Mark's numbers.

We need to be focusing less on artificial tests such as VUPS.COM and more
on the real-world tests like the ones that Mark is running.

Simon.

--
Simon Clubley, clubley@remove_me.eisner.decus.org-Earth.UFP
Walking destinations on a map are further away than they appear.

On 2023-04-28, Volker Halle <volker_halle@hotmail.com> wrote:
>
> We can continue the VUPS discussion here, but should discuss the KERNEL mode question in the VSI forum.
>

Why ? The Kernel mode overhead is the real issue that needs discussing,
not some artificial CPU speed test.

Simon.

--
Simon Clubley, clubley@remove_me.eisner.decus.org-Earth.UFP
Walking destinations on a map are further away than they appear.

On 28/4/2023 4:11 pm, Volker Halle wrote:
> Mark,
>
> there are at least 2 variants of the 'VUPS' procedure. And some of these variants have my name in them.
>
> Stromasys variant: I have been using this variant since about 19 years in CHARON-VAX and CHARON-AXP migration projects. This procedure was supposed to report the 'correct VUPS' values as documented for the older real VAX and Alpha systems:
>
> VUPS numbers for 'real systems' were publicized by DEC, but most of these documents have disappeared from the WEB.
>
> Here are some examples: http://www.vaxmacro.de/vvcc.html
>
> I had to modify the Stromasys VUPS procedure (called SRI_VUPS.COM at that time, SRI = Software Resources International, which later became Stromasys), because it would enter an endless loop on the faster Alphas (> 1 GHz). And this modification found it's way into the other variants below.
>
> FreeAXP variant: http://www.migrationspecialties.com/Downloads/VUPS.txt
> EmuVM variant: https://emuvm.com/download/vups-com-benchmark/
>
> The above 2 are strongly related (as can be seen by this line: $ cpu_multiplier = 10 ! VAX = 10 - Alpha/AXP = 40). And there is no code in these procedures to check the architecture they're running on and adapt the CPU_MULTIPLIER ! So the VUPS results produced can not be directly compared between the Stromasys variant and the other two - although there should be a linear relation between them. The DCL code in the 'inner loop' is the same for all these variants. Pure integer performance on one single CPU.
>
> The idea of this procedure was the ability to compare 'VUPS' values between hardware systems and their emulated counterparts. Simple DCL procedure, could be mailed and run on the target system. No compiling required. When OpenVMS on I64 came around, I was interested in also calculating the 'VUPS' values for those systems. I used the runtime of the PRIME_SIEVE program to 'calibrate' the CPU_MODIFIER value in the SRI_VUPS.COM variant, so it would report VUPS values, which allowed to compare the VAX, Alpha, I64 and now x86-64 CPUs under OpenVMS.
>
> So whenever someone provides VUPS values, they need to also document, which procedure has been used to collect them - as John has done by calling your procedure WASD_VUPS.COM
>
> We can continue the VUPS discussion here, but should discuss the KERNEL mode question in the VSI forum.
>
> Volker.

This is slightly off the topic of KERNEL mode consumption on X86 but ...

I have put all three into a ZIP archive at

https://wasd.vsm.com.au/other/vups_230428a.zip
https://wasd.vsm.com.au/other/vups_230428a.txt

for anyone who would like to experiment.

The comparative results across a number of platforms I have access to
are as follows (extraneous white-space removed).

|$ @wasd_vups.com
|HP rx2660 (1.40GHz/6.0MB) with 4 CPU and 14335MB running VMS V8.4-2L3
|INFO: Preventing endless loop (10$) on fast CPUs
|Approximate System VUPs Rating : 469.0 ( min: 465.8 max: 470.6 )
|$ @migs_vups.com
|INFO: Preventing endless loop (10$) on fast CPUs
|Approximate System VUPs Rating : 456.5 ( min: 453 max: 458 )
|$ @emuvm_vups.com
|Approximate System VUPs Rating : 458.0 ( min: 458.0 max: 458.0 )

|$ @wasd_vups.com
|Digital Personal WorkStation with 1 CPU and 1536MB running VMS V8.4-2L1
|Approximate System VUPs Rating : 151.4 ( min: 151.0 max: 151.8 )
|$ @migs_vups.com
|Approximate System VUPs Rating : 144.5 ( min: 144 max: 144 )
|$ @emuvm_vups.com
|Approximate System VUPs Rating : 145.0 ( min: 144.8 max: 145.4 )

|! EISNER
|$ @WASD_VUPS.COM
|AlphaServer DS20 500 MHz with 2 CPU and 1536MB running VMS V8.4-2L2
|Approximate System VUPs Rating : 243.5 ( min: 238.6 max: 247.6 )
|$ @migs_VUPS.COM
|Approximate System VUPs Rating : 236.5 ( min: 234 max: 239 )
|$ @emuvm_VUPS.COM
|Approximate System VUPs Rating : 239.5 ( min: 238.2 max: 241.8 )

|! Dell Optiplex 9020 4 core i7 3.4Ghz 16GB Win 10 Pro 22H2
|$ @wasd_vups.com
|innotek GmbH VirtualBox with 2 CPU and 7574MB running VMS V9.2
|Approximate System VUPs Rating : 282.2 ( min: 281.4 max: 282.6 )
|$ @migs_vups.com
|INFO: Preventing endless loop (10$) on fast CPUs
|Approximate System VUPs Rating : 275.3 ( min: 273 max: 276 )
|$ @emuvm_vups.com
|Approximate System VUPs Rating : 275.8 ( min: 275.8 max: 275.8 )

|! BXNUC10i7FNH4 6 core i7 1.10GHz 32GB
|VSMX86$ @wasd_vups.com
|innotek GmbH VirtualBox with 2 CPU and 7936MB running VMS E9.2-1
|Approximate System VUPs Rating : 645.4 ( min: 642.4 max: 647.0 )
|$ @migs_vups.com
|INFO: Preventing endless loop (10$) on fast CPUs
|Approximate System VUPs Rating : 641.2 ( min: 641 max: 641 )
|$ @emuvm_vups.com
|Approximate System VUPs Rating : 636.3 ( min: 635.4 max: 638.2 )

Observations ... the VUPS.COM numbers all seem comparable between
versions of the procedure. The WASD_VUPS.COM numbers perhaps marginally
higher than the other two.

When the adjustment re X86 per Volker's post are uncommented

|$!!! if f$getsyi("arch_name").eqs."x86_64" then $ cpu_multiplier=8

the numbers drop by some 20%.

|! Dell Optiplex 9020 4 core i7 3.4Ghz 16GB Win 10 Pro 22H2
|$ @wasd_vups.com
|innotek GmbH VirtualBox with 2 CPU and 7574MB running VMS V9.2
|Approximate System VUPs Rating : 225.6 ( min: 225.6 max: 225.6 )

|! Dell Optiplex 9020 4 core i7 3.4Ghz 16GB Win 10 Pro 22H2
|$ @wasd_vups.com
|innotek GmbH VirtualBox with 2 CPU and 7936MB running VMS E9.2-1
|INFO: Preventing endless loop (10$) on fast CPUs
|Approximate System VUPs Rating : 517.8 ( min: 515.8 max: 518.8 )

And now back to the main show, KERNEL mode numbers

https://wasd.vsm.com.au/info-WASD/2023/0077

--
Anyone, who using social-media, forms an opinion regarding anything
other than the relative cuteness of this or that puppy-dog, needs
seriously to examine their critical thinking.

On 4/28/23 8:33 AM, Mark Daniel wrote:
> .
> .
> .
> This is slightly off the topic of KERNEL mode consumption on X86 but ...
>
> I have put all three into a ZIP archive at
>
>    https://wasd.vsm.com.au/other/vups_230428a.zip
>    https://wasd.vsm.com.au/other/vups_230428a.txt
>
> for anyone who would like to experiment.
>
> The comparative results across a number of platforms I have access to
> are as follows (extraneous white-space removed).
> .
> .
> .

Some results from my own systems:

|$ @wasd_vups
|AlphaServer DS15 with 1 CPU and 2048MB running VMS V8.4-2L2
|INFO: Preventing endless loop (10$) on fast CPUs
|Approximate System VUPs Rating : 468.7 ( min: 467.4 max: 470.6 )
| |$ @migs_vups
|INFO: Preventing endless loop (10$) on fast CPUs
|Approximate System VUPs Rating : 451.2 ( min: 448 max: 452 )
| |$ @emuvm_vups
|INFO: Preventing endless loop (10$) on fast CPUs
|Approximate System VUPs Rating : 452.0 ( min: 452.0 max: 452.0 )
| |$ run bogomips_axp
|AlphaServer DS15 1 CPU 2048MB V8.4-2L2
|Calibrating delay loop.. ok - 994.20 BogoMips

|$ @wasd_vups
|HP rx2620 (1.40GHz/6.0MB) with 4 CPU and 6143MB running VMS V8.4-2L3
|INFO: Preventing endless loop (10$) on fast CPUs
|Approximate System VUPs Rating : 488.1 ( min: 487.2 max: 488.8 )
| |$ @migs_vups
|INFO: Preventing endless loop (10$) on fast CPUs
|Approximate System VUPs Rating : 473.2 ( min: 472 max: 473 )
| |$ @emuvm_vups
|INFO: Preventing endless loop (10$) on fast CPUs
|Approximate System VUPs Rating : 473.8 ( min: 473.8 max: 473.8 )
| |$ run bogomips_i64
|HP rx2620 (1.40GHz/6.0MB) 4 CPUs 6143MB V8.4-2L3
|Calibrating delay loop.. ok - 692.73 BogoMips

This next system is a virtual guest of an ESXi host.
The host is a DL380 Gen9 with Intel E5-2660 v3 CPUs at 2.60GHz
|$ @wasd_vups
|VMware, Inc. VMware7,1 with 4 CPU and 16124MB running VMS E9.2-1
|Approximate System VUPs Rating : 509.6 ( min: 496.6 max: 516.2 )
| |$ @migs_vups
|Approximate System VUPs Rating : 504.2 ( min: 492 max: 508 )
| |$ @emuvm_vups
|Approximate System VUPs Rating : 503.8 ( min: 486.2 max: 510.4 )
| |$ run bogomips_x86
|VMware, Inc. VMware7,1 4 CPUs 16124MB E9.2-1
|Calibrating delay loop.. ok - 909.95 BogoMips

It was necessary to compile bogomips /noopt on x86. Otherwise, the
delay loop was optimized away. (For fairness, /noopt was used on the
other platforms as well).

If I knew what arguments to give to the prime_sieve program, I would
have included that as well.

Mark Berryman

On Thu, 2023-04-27 at 20:19 -0400, fsword007@gmail.com wrote:
> $ @vups2
> innotek GmbH VirtualBox with 2 CPU and 7936MB running VMS E9.2-1
> INFO: Preventing endless loop (10$) on fast CPUs
>  
> Approximate System VUPs Rating : 414.8 ( min: 408.4 max: 418.0 )

Where did you get your vups2 from?

I neglected to mention what my system is: AMD Threadripper 2970wx (48
logical cores) with 128GB ECC RAM. VM runs on a spinning rust RAID
array, although plans are afoot to replace that with NVME RAID.
--
Tactical Nuclear Kittens

On Thu, 2023-04-27 at 23:26 -0500, John H Reinhardt wrote:
> Approximate System VUPs Rating : 693.1 ( min: 658.2 max: 705.6 )

That is _quick_!
--
Tactical Nuclear Kittens

Mark Berryman schrieb am Freitag, 28. April 2023 um 19:41:23 UTC+2:

> If I knew what arguments to give to the prime_sieve program, I would
> have included that as well.

Mark,

I typically run PRIME_SIEVE to calculate prime numbers between 1 and 10^10, so just give it 10000000000

$ ps :==$dev:<dir>prime_sieve.exe
$ ps 10000000000
# {1 <= primes <= 10000000000} = 455052511

and report F$GETSYI("CPUTIM") difference for the execution.

Volker.

On 4/28/23 12:03 PM, Volker Halle wrote:
> Mark Berryman schrieb am Freitag, 28. April 2023 um 19:41:23 UTC+2:
>
>> If I knew what arguments to give to the prime_sieve program, I would
>> have included that as well.
>
> Mark,
>
> I typically run PRIME_SIEVE to calculate prime numbers between 1 and 10^10, so just give it 10000000000
>
> $ ps :==$dev:<dir>prime_sieve.exe
> $ ps 10000000000
> # {1 <= primes <= 10000000000} = 455052511
>
> and report F$GETSYI("CPUTIM") difference for the execution.

Ah. In that case:

Alpha: 7577
IA64: 6813
x86_64: 3306

Mark Berryman

Den 2023-04-28 kl. 14:22, skrev Simon Clubley:
> On 2023-04-28, Volker Halle <volker_halle@hotmail.com> wrote:
>>
>> We can continue the VUPS discussion here, but should discuss the KERNEL mode question in the VSI forum.
>>
>
> Why ? The Kernel mode overhead is the real issue that needs discussing,
> not some artificial CPU speed test.
>
> Simon.
>

Why? That should be very obvious.

There is already an ongoing diskussion over at the Forum about this.
The posts and code posted are much eiser to read.
Much better thread history.

If the kernel overhead is viewed as "the real problem", it is better
disussed on the Forum, not here...

On Friday, April 28, 2023 at 10:33:49 AM UTC-4, Mark Daniel wrote:
> I have put all three into a ZIP archive at
>
> https://wasd.vsm.com.au/other/vups_230428a.zip
> https://wasd.vsm.com.au/other/vups_230428a.txt
>
> for anyone who would like to experiment.

This is on an EmuVM Alpha with a FreeBSD 12.4 host OS with dual E5-2643 v3 CPUs:

SERVER::$ unzip vups_230428a.zip
Archive: SYS$SYSDEVICE:[SYSPROG.TERRY]VUPS_230428A.ZIP;1
inflating: emuvm_vups.com
inflating: migs_vups.com
inflating: wasd_vups.com
SERVER::$ @emuvm_vups
Approximate System VUPs Rating : 825.4 ( min: 774.6 max: 845.8 )
SERVER::$ @migs_vups
INFO: Preventing endless loop (10$) on fast CPUs
INFO: Preventing endless loop (10$) on fast CPUs
Approximate System VUPs Rating : 832.5 ( min: 799 max: 850 )
SERVER::$ @wasd_vups
AlphaServer DS10 616 MHz with 1 CPU and 2048MB running VMS V8.4-2L1
INFO: Preventing endless loop (10$) on fast CPUs
Approximate System VUPs Rating : 889.2 ( min: 811.0 max: 925.2 )

On 4/28/2023 10:07 PM, terry-...@glaver.org wrote:
> On Friday, April 28, 2023 at 10:33:49 AM UTC-4, Mark Daniel wrote:
>> I have put all three into a ZIP archive at
>>
>> https://wasd.vsm.com.au/other/vups_230428a.zip
>> https://wasd.vsm.com.au/other/vups_230428a.txt
>>
>> for anyone who would like to experiment.
>
> This is on an EmuVM Alpha with a FreeBSD 12.4 host OS with dual E5-2643 v3 CPUs:
>
> SERVER::$ unzip vups_230428a.zip
> Archive: SYS$SYSDEVICE:[SYSPROG.TERRY]VUPS_230428A.ZIP;1
> inflating: emuvm_vups.com
> inflating: migs_vups.com
> inflating: wasd_vups.com
> SERVER::$ @emuvm_vups
>
> Approximate System VUPs Rating : 825.4 ( min: 774.6 max: 845.8 )
> SERVER::$ @migs_vups
> INFO: Preventing endless loop (10$) on fast CPUs
> INFO: Preventing endless loop (10$) on fast CPUs
>
> Approximate System VUPs Rating : 832.5 ( min: 799 max: 850 )
> SERVER::$ @wasd_vups
> AlphaServer DS10 616 MHz with 1 CPU and 2048MB running VMS V8.4-2L1
> INFO: Preventing endless loop (10$) on fast CPUs
>
> Approximate System VUPs Rating : 889.2 ( min: 811.0 max: 925.2 )

By comparison, your DS10 is a hot rod vs my two. I have a "stock" DS10 617MHz and another that is a DS10L 617MHz mainboard I transplanted into a regular DS10 chassis. Both VUP out below half of your figure.

Stock DS10:

$ @wasd_vups
AlphaServer DS10 617 MHz with 1 CPU and 2048MB running VMS V8.4-2L1
INFO: Preventing endless loop (10$) on fast CPUs

Approximate System VUPs Rating : 331.4 ( min: 329.4 max: 332.8 )

Upgraded DS10:
$ @wasd_vups
AlphaServer DS10L 617 MHz with 1 CPU and 2048MB running VMS V8.4-2L1
INFO: Preventing endless loop (10$) on fast CPUs

Approximate System VUPs Rating : 325.8 ( min: 324.6 max: 327.0 )

I wonder what the difference is.

John H. Reinhardt

Volker Halle schrieb am Donnerstag, 27. April 2023 um 19:49:13 UTC+2:

Here are the correct commands for the PCS$SDA PC sampling SDA extension - sorry about that ;-(

$ ANA/SYS
SDA> READ/EXEC ! get better symbolization (routine names)
SDA> PCS LOAD ! PC sampling SDA Execlet
SDA> PCS START TRACE
....
SDA> PCS STOP TRACE
SDA> PCS SHOW TRACE/STAT/MODE=KERNEL
....
SDA> PCS UNLOAD
SDA> EXIT
SDA> PCS SHOW TRACE/STAT/MODE=KERNEL will show the PC values seen most often during the trace period (sorted by decreasing no. of occurences) and it will symbolize the PC values as routine names or execlet names. This might give you an idea, which kind of kernel mode code is running how often.

Volker.

On Saturday, April 29, 2023 at 12:07:16 AM UTC-4, John H Reinhardt wrote:
> By comparison, your DS10 is a hot rod vs my two. I have a "stock" DS10 617MHz and another that is a DS10L 617MHz mainboard I transplanted into a regular DS10 chassis. Both VUP out below half of your figure.
>
> I wonder what the difference is.

Mine is emulated via AlphaVM on a Dell Poweredge R730.

It would be interesting to run the same tests on native X86-64 VMS on the same hardware.

On 4/29/2023 12:26 AM, terry-...@glaver.org wrote:
> On Saturday, April 29, 2023 at 12:07:16 AM UTC-4, John H Reinhardt wrote:
>> By comparison, your DS10 is a hot rod vs my two. I have a "stock" DS10 617MHz and another that is a DS10L 617MHz mainboard I transplanted into a regular DS10 chassis. Both VUP out below half of your figure.
>>
>> I wonder what the difference is.
>
> Mine is emulated via AlphaVM on a Dell Poweredge R730.
>
Doh! I missed that.

> It would be interesting to run the same tests on native X86-64 VMS on the same hardware.

Yes it would

Running the various VUPS procedures on my VSI OpenVMS x86-64 E9.2-1 system on VMware Workstation Player 16 on an i5-9600K CPU @3.7 GHz:

$ @wasd_vups.com
VMware, Inc. VMware7,1 with 2 CPU and 3836MB running VMS E9.2-1
INFO: Preventing endless loop (10$) on fast CPUs
Approximate System VUPs Rating : 757.6 ( min: 757.6 max: 757.6 )

$ @migs_vups.com
INFO: Preventing endless loop (10$) on fast CPUs
Approximate System VUPs Rating : 747.9 ( min: 745 max: 753 )

$ @emuvm_vups.com
INFO: Preventing endless loop (10$) on fast CPUs
Approximate System VUPs Rating : 749.2 ( min: 745.2 max: 753.4 )

It should be noted that the following VUPS procedure from Stromasys has been 'calibrated' based on the relative PRIME_SIEVE integer CPU performance (see link above) data to deliver CPU performance data comparable across all 4 different OpenVMS architectures (VAX, Alpha, I64 and x86-64).

$ @SRI_VUPS_NEW.COM
Approximate System VUPs Rating : 3706.0 ( min: 3700 max: 3717 )

PRIME_SIEVE - calculating no. of primes in 10^10 ...
# {1 <= primes <= 10000000000} = 455052511
Calculated no. of primes in 10^10 - end=18:04:31.42 start=18:04:12.11
Delta time is 00:00:19.31 - Delta CPUTIM is 1924 ticks

Volker.

terry- schrieb am Samstag, 29. April 2023 um 07:26:14 UTC+2:

> It would be interesting to run the same tests on native X86-64 VMS on the same hardware.

Terry,

I would not expect much difference when running these VUPS tests natively on physical hardware. It's mostly CPU bound and this is 'virtualization' not 'emulation' and the instructions are executed unchanged on the physical processor. IOs would need some intervention by the hypervisor and that could make a difference.

Volker.

On Saturday, April 29, 2023 at 12:17:31 PM UTC-4, Volker Halle wrote:
> terry- schrieb am Samstag, 29. April 2023 um 07:26:14 UTC+2:
>
> > It would be interesting to run the same tests on native X86-64 VMS on the same hardware.
> Terry,
>
> I would not expect much difference when running these VUPS tests natively on physical hardware. It's mostly CPU bound and this is 'virtualization' not 'emulation' and the instructions are executed unchanged on the physical processor. IOs would need some intervention by the hypervisor and that could make a difference.

But my emulator is providing Alpha hardware emulation (per the "DS10" line in my VUPS output). What I'm interested in seeing is how VMS x86-64 performs on this Dell server, compared to the additional overhead of providing a complete emulated DS10.

On 4/29/2023 2:53 PM, terry-...@glaver.org wrote:

> But my emulator is providing Alpha hardware emulation (per the "DS10" line in
> my VUPS output). What I'm interested in seeing is how VMS x86-64 performs on
> this Dell server, compared to the additional overhead of providing a complete
> emulated DS10.

Any performance numbers for VMS on X86_64 are not *that* interesting until it's
built with native optimizing compilers.

Native compilers are becoming available, but the build process will take some
time to convert to a native build.

I do not think (but am not sure) that V9.2-2 (scheduled for near the end of the
year) would be a native build. Early builds of that version are being done
with cross compilers.

--

--- Rob

On Friday, April 28, 2023 at 8:22:29 AM UTC-4, Simon Clubley wrote:
> On 2023-04-28, Volker Halle <volker...@hotmail.com> wrote:
> >
> > We can continue the VUPS discussion here, but should discuss the KERNEL mode question in the VSI forum.
> >
> Why ? The Kernel mode overhead is the real issue that needs discussing,
> not some artificial CPU speed test.
> Simon.

I made a wrapper for the prime_sieve.c V2 program (making it a callable function and redirecting its calls to printf)
so I could experiment with making it multi-threaded. My DS10L/466 Alpha can do the 10^9 test in 20.1 seconds if
using a single worker that assumes only 1 worker. If you remove that assumption (create a 2nd thread), the time
goes up to 26.8 seconds due to the ~500,000 calls to pthread_getspecific. My emulated DS20, which is emulating
2 processors, goes from 51.7 seconds to 57.6 for the same experiment. A fair amount of overhead, but multiple
CPUs will improve the final run time.

On my X86 hardware (i5-1235U), the single worker case takes 3.4 seconds while adding in the pthread_calls increases
the run time to 37.6 seconds! In the first case, monitor shows 100% user mode while the second case show 50%
kernel mode.

terry- schrieb am Samstag, 29. April 2023 um 20:53:39 UTC+2:

> But my emulator is providing Alpha hardware emulation (per the "DS10" line in my VUPS output). What I'm interested in seeing is how VMS x86-64 performs on
> this Dell server, compared to the additional overhead of providing a complete emulated DS10.

Terry,

if you can, install a supported Hypervisor on this DELL server under FreeBSD 12.4 host OS, create a VM, install VSI OpenVMS x86-64 and see for yourself.

Volker.

David Jones schrieb am Samstag, 29. April 2023 um 21:56:27 UTC+2:

David,

thanks for this experiment. It shows the current huge overhead, when introducing kernel mode calls. Let's be fair: VSI is still concentrating on functionality vs. performance for OpenVMS x86-64. So we should wait for the native build of the OS - as indicated by Rob.

Volker.

On Sunday, April 30, 2023 at 12:26:41 AM UTC-4, Volker Halle wrote:
> if you can, install a supported Hypervisor on this DELL server under FreeBSD 12.4 host OS, create a VM, install VSI OpenVMS x86-64 and see for yourself.

I haven't received access to the kits / licenses yet.

When I do, I'll do it on a different R730 (same CPUS / memory) as this box is hosting my AlphaVM instance which I depend on for email and is also serving various VMS freeware.

Just to add to the other stats presented...

VirtualBox 7.0 on a HP ProBook 450 G8 laptop having:
11th Gen Intel(R) Core(TM) i5-1135G7 @ 2.40GHz 2.42 GHz
running Win10 Pro.

COMs are from the ZIP kit from Daniel.

Removed that "preventing" output and lining up the numbers...

$ @wasd_vups.com
innotek GmbH VirtualBox with 2 CPU and 3840MB running VMS E9.2-1
Approximate System VUPs Rating : 778.2 ( min: 770.2 max: 783.4 )

$ @migs_vups.com
Approximate System VUPs Rating : 766.0 ( min: 766 max: 766 )

$ @emuvm_vups.com
Approximate System VUPs Rating : 765.2 ( min: 761.6 max: 770.2 )

Our current DS20e Alphas:

$ @WASD_VUPS.COM
COMPAQ AlphaServer DS20E 666 MHz with 1 CPU and 4096MB running VMS V8.4-2L2
Approximate System VUPs Rating : 354.4 ( min: 354.4 max: 354.4 )

$ @MIGS_VUPS.COM
Approximate System VUPs Rating : 354.3 ( min: 352 max: 356 )

$ @EMUVM_VUPS.COM
Approximate System VUPs Rating : 354.0 ( min: 353.6 max: 355.4 )

Interesting that these are much more close to each other then
on the x86 test.

I also noted that the WASD version did *not* print the "Preventing
endless loop" message on the DS20e, only the MIGS version did...

Interesting that this i5 2.4 GHz laptop gave better numbers then the
i5 3.7 GHz system from Volker below...

Regards, Jan-Erik.

Den 2023-04-29 kl. 18:07, skrev Volker Halle:
> Running the various VUPS procedures on my VSI OpenVMS x86-64 E9.2-1 system on VMware Workstation Player 16 on an i5-9600K CPU @3.7 GHz:
>
> $ @wasd_vups.com
> VMware, Inc. VMware7,1 with 2 CPU and 3836MB running VMS E9.2-1
> INFO: Preventing endless loop (10$) on fast CPUs
> Approximate System VUPs Rating : 757.6 ( min: 757.6 max: 757.6 )
>
> $ @migs_vups.com
> INFO: Preventing endless loop (10$) on fast CPUs
> Approximate System VUPs Rating : 747.9 ( min: 745 max: 753 )
>
> $ @emuvm_vups.com
> INFO: Preventing endless loop (10$) on fast CPUs
> Approximate System VUPs Rating : 749.2 ( min: 745.2 max: 753.4 )
>
> It should be noted that the following VUPS procedure from Stromasys has been 'calibrated' based on the relative PRIME_SIEVE integer CPU performance (see link above) data to deliver CPU performance data comparable across all 4 different OpenVMS architectures (VAX, Alpha, I64 and x86-64).
>
> $ @SRI_VUPS_NEW.COM
> Approximate System VUPs Rating : 3706.0 ( min: 3700 max: 3717 )
>
> PRIME_SIEVE - calculating no. of primes in 10^10 ...
> # {1 <= primes <= 10000000000} = 455052511
> Calculated no. of primes in 10^10 - end=18:04:31.42 start=18:04:12.11
> Delta time is 00:00:19.31 - Delta CPUTIM is 1924 ticks
>
> Volker.

Jan-Erik Söderholm schrieb am Sonntag, 30. April 2023 um 13:15:12 UTC+2:

> Interesting that these are much more close to each other then
> on the x86 test.

DS20Es are physical systems and there is probably less interference than from the VM hypervisor and host OS.

> I also noted that the WASD version did *not* print the "Preventing
> endless loop" message on the DS20e, only the MIGS version did...

I had to add these checks on 'faster' CPUs (Alpha 1 GHz or higher) to prevent endless loops, if the no. of executions of the internal DCL loop were not enough to at least use 10 ms (1 tick) worth of CPUTIM.

> Interesting that this i5 2.4 GHz laptop gave better numbers then the
> i5 3.7 GHz system from Volker below...

the various VUPS results on your i5-1135G7 @2.4 GHz are only about 2,7% higher than on my i5-9600K @3.7 GHz and there are certanly other components involved in this test than just the plain CPU speed.

I normally use PassMarks to compare CPUs: https://www.cpubenchmark.net/ and the Single thread rating values of your (2710) and my (2759) CPUs are very similar (mine is only rated 1,8% faster). So I find it amazing, that these VUPS values are actually so close ;-)

Volker.

Den 2023-04-30 kl. 13:52, skrev Volker Halle:
> Jan-Erik Söderholm schrieb am Sonntag, 30. April 2023 um 13:15:12 UTC+2:
>
>> Interesting that these are much more close to each other then
>> on the x86 test.
>
> DS20Es are physical systems and there is probably less interference than from the VM hypervisor and host OS.
>
>> I also noted that the WASD version did *not* print the "Preventing
>> endless loop" message on the DS20e, only the MIGS version did...
>
> I had to add these checks on 'faster' CPUs (Alpha 1 GHz or higher) to prevent endless loops, if the no. of executions of the internal DCL loop were not enough to at least use 10 ms (1 tick) worth of CPUTIM.
>
>> Interesting that this i5 2.4 GHz laptop gave better numbers then the
>> i5 3.7 GHz system from Volker below...
>
> the various VUPS results on your i5-1135G7 @2.4 GHz are only about 2,7% higher than on my i5-9600K @3.7 GHz and there are certanly other components involved in this test than just the plain CPU speed.
>
> I normally use PassMarks to compare CPUs: https://www.cpubenchmark.net/ and the Single thread rating values of your (2710) and my (2759) CPUs are very similar (mine is only rated 1,8% faster). So I find it amazing, that these VUPS values are actually so close ;-)
>
> Volker.

OK. I might have remembered the "endless loop" message from an
DS25 I once had.

And yes, there are too many "unknowns" to really compare things here...

Anyway, it's up and running, and that is fun enough. Will probably
allocate some unused laptop as the "VMS system".

Have a box of Lantronix ETS8 terminal servers that I was going to
verify/reset, and that is also easier with an VMS system available.

Now, I'm just holding my breath until I have a Cobol compiler to
play with... :-)