"2321" packages to upgrade . . . on reboot another "204" yet to go?? Why???

Hi
Can you show the output, are you in $HOME?


cd
pwd

RAID1 on rotating rust (2 1TB Seagates) and 5 year old 2 core 4 thread 3.4 GHz i3 CPU with same conditions: all packages downloaded to the LAN server in advance, so zypper processing 1382 TW20220629 packages only, in 25m32s real, 2m34s user, 1m23s sys.

@malcolmlewis:

Spent all night trying to figure out if I am in $HOME . . . but, I’m assuming that if you launch the Terminal freshly, that is where you are??? I was I believe, in “home” before I tried to run those commands . . . and “onemkl” directory is extracted “in the Home Directory” . . . but did not respond, to the commands provided, etc.

butoh-time@linux-fl:~> cd
butoh-time@linux-fl:~> pwd
/home/butoh-time
butoh-time@linux-fl:~> cd /home/butoh-time
butoh-time@linux-fl:~>

Note: some names have been changed to protect the innocent . . . .

Hi
Ahh it’s benchmark;


ls prim* l-one*

l-onemklbench-p-2022-0-2-84.tgz


l-onemklbench-p-2022-0-2-84:
benchmarks_2022.0.2


prime95:
libgmp.so  libgmp.so.10  libgmp.so.10.4.1  license.txt  mprime  readme.txt  stress.txt  undoc.txt  whatsnew.txt

So would be;


cd benchmarks_2022.0.2//linux/mkl/benchmarks/linpack

@malcolmlewis:

Yes, that is working . . . the stressing is being done as we speak, so thanks for that . . . right now we are up to “18000” and it’s all showing as “pass” . . . .

**erlangen:~ #** time echo "scale=5000; 4*a(1)" | bc -l             

real    0m10.541s 
user    0m10.539s 
sys     0m0.001s 

Is this command from Karl something that just tests up to “5000” and then shows the results?? I’m running out of time and it seems to be taking longer on the
18000 and above testing??

My newest zipped along at a pretty good clip on TW20220630, but with considerable delays at rpcbind and POST. 16GB RAM, 6 cores, 12 threads, 2.4GHz, 65W TDP, i5-11400: 995 packages from LAN/NFS in 2m08s real, 0m47s user, 0m30s sys.

karl@6700K:~/Downloads/benchmarks_2021.2.0/linux/mkl/benchmarks/linpack> ./runme_xeon64 
This is a SAMPLE run script for running a shared-memory version of 
Intel(R) Distribution for LINPACK* Benchmark. Change it to reflect 
the correct number of CPUs/threads, problem input files, etc.. 
*Other names and brands may be claimed as the property of others. 
So 3. Jul 13:40:23 CEST 2022 
Sample data file lininput_xeon64. 

Current date/time: Sun Jul  3 13:40:23 2022 

CPU frequency:    4.199 GHz 
Number of CPUs: 1 
Number of cores: 4 
Number of threads: 4 

Parameters are set to: 

Number of tests: 15 
Number of equations to solve (problem size) : 1000  2000  5000  10000 15000 18000 20000 22000 25000 26000 27000 30000 35000 40000 45000 
Leading dimension of array                  : 1000  2000  5008  10000 15000 18008 20016 22008 25000 26000 27000 30000 35000 40000 45000 
Number of trials to run                     : 4     2     2     2     2     2     2     2     2     2     1     1     1     1     1     
Data alignment value (in Kbytes)            : 4     4     4     4     4     4     4     4     4     4     4     1     1     1     1     

Maximum memory requested that can be used=16200901024, at the size=45000 

=================== Timing linear equation system solver =================== 

Size   LDA    Align. Time(s)    GFlops   Residual     Residual(norm) Check 
1000   1000   4      0.006      117.7213 1.022959e-12 3.033181e-02   pass 
1000   1000   4      0.005      126.1908 1.022959e-12 3.033181e-02   pass 
1000   1000   4      0.004      152.2414 1.022959e-12 3.033181e-02   pass 
1000   1000   4      0.004      161.0816 1.022959e-12 3.033181e-02   pass 
2000   2000   4      0.038      140.4579 5.619838e-12 4.375464e-02   pass 
2000   2000   4      0.036      146.4662 5.619838e-12 4.375464e-02   pass 
5000   5008   4      0.445      187.2600 2.548040e-11 3.392018e-02   pass 
5000   5008   4      0.443      188.2553 2.548040e-11 3.392018e-02   pass 
10000  10000  4      3.240      205.8410 1.054555e-10 3.553909e-02   pass 
10000  10000  4      3.241      205.7734 1.054555e-10 3.553909e-02   pass 
15000  15000  4      10.624     211.8316 2.368669e-10 3.581348e-02   pass 
15000  15000  4      10.610     212.1047 2.368669e-10 3.581348e-02   pass 
18000  18008  4      18.150     214.2523 3.162348e-10 3.349350e-02   pass 
18000  18008  4      18.166     214.0648 3.162348e-10 3.349350e-02   pass 
20000  20016  4      24.739     215.6194 3.807211e-10 3.257923e-02   pass 
20000  20016  4      24.735     215.6508 3.807211e-10 3.257923e-02   pass 
22000  22008  4      32.748     216.7986 4.590843e-10 3.258820e-02   pass 
22000  22008  4      32.756     216.7424 4.590843e-10 3.258820e-02   pass 
25000  25000  4      47.808     217.9128 5.770316e-10 3.184866e-02   pass 
25000  25000  4      47.753     218.1608 5.770316e-10 3.184866e-02   pass 
26000  26000  4      53.590     218.6746 6.257559e-10 3.196386e-02   pass 
26000  26000  4      53.542     218.8676 6.257559e-10 3.196386e-02   pass 
27000  27000  4      59.979     218.8002 5.721172e-10 2.712944e-02   pass 
30000  30000  1      82.055     219.3875 7.350489e-10 2.829664e-02   pass 
35000  35000  1      128.858    221.8386 1.150995e-09 3.248513e-02   pass 
40000  40000  1      191.738    222.5425 1.375035e-09 2.988233e-02   pass 
45000  45000  1      272.779    222.7224 1.859673e-09 3.192513e-02   pass 

Performance Summary (GFlops) 

Size   LDA    Align.  Average  Maximal 
1000   1000   4       139.3088 161.0816 
2000   2000   4       143.4620 146.4662 
5000   5008   4       187.7577 188.2553 
10000  10000  4       205.8072 205.8410 
15000  15000  4       211.9682 212.1047 
18000  18008  4       214.1586 214.2523 
20000  20016  4       215.6351 215.6508 
22000  22008  4       216.7705 216.7986 
25000  25000  4       218.0368 218.1608 
26000  26000  4       218.7711 218.8676 
27000  27000  4       218.8002 218.8002 
30000  30000  1       219.3875 219.3875 
35000  35000  1       221.8386 221.8386 
40000  40000  1       222.5425 222.5425 
45000  45000  1       222.7224 222.7224 

Residual checks PASSED 

End of tests 

Done: So 3. Jul 14:00:10 CEST 2022 
karl@6700K:~/Downloads/benchmarks_2021.2.0/linux/mkl/benchmarks/linpack> 
**
6700K:~ #** inxi -zSMmC 
**System:**
  **Kernel:** 5.18.6-1-default **arch:** x86_64 **bits:** 64 **Console:** pty pts/1 
    **Distro:** openSUSE Tumbleweed 20220701 
**Machine:**
  **Type:** Desktop **Mobo:** ASRock **model:** Z170 Pro4S **serial:** <filter> 
    **UEFI:** American Megatrends **v:** P7.50 **date:** 01/23/2018 
**Memory:**
  **RAM:****total:** 31.04 GiB **used:** 2.99 GiB (9.6%) 
  **Array-1:****capacity:** 64 GiB **slots:** 4 **EC:** None 
  **Device-1:** ChannelA-DIMM0 **type:** no module installed 
  **Device-2:** ChannelA-DIMM1 **type:** DDR4 **size:** 16 GiB **speed:** 2133 MT/s 
  **Device-3:** ChannelB-DIMM0 **type:** no module installed 
  **Device-4:** ChannelB-DIMM1 **type:** DDR4 **size:** 16 GiB **speed:** 2133 MT/s 
**CPU:**
  **Info:** quad core **model:** Intel Core i7-6700K **bits:** 64 **type:** MT MCP **cache:**
    **L2:** 1024 KiB 
  **Speed (MHz):****avg:** 800 **min/max:** 800/4200 **cores:****1:** 800 **2:** 800 **3:** 800 
    **4:** 800 **5:** 800 **6:** 800 **7:** 800 **8:** 800 
**6700K:~ #** 

What is yours?

Hi


inxi -SMmC 

System:
  Host: grover Kernel: 5.18.6-1-default arch: x86_64 bits: 64 Desktop: GNOME
    v: 42.2 Distro: openSUSE Tumbleweed 20220702
Machine:
  Type: Desktop System: Hewlett-Packard product: HP Z440 Workstation v: N/A
    serial: <filter>
  Mobo: Hewlett-Packard model: 212B v: 1.01 serial: N/A
    UEFI: Hewlett-Packard v: M60 v02.58 date: 12/07/2021
Memory:
  RAM: total: 62.71 GiB used: 3.84 GiB (6.1%)
  Array-1: capacity: 512 GiB note: check slots: 8 EC: Multi-bit ECC
  Device-1: CPU0-DIMM1 type: DDR4 size: 8 GiB speed: 2133 MT/s
  Device-2: CPU0-DIMM2 type: DDR4 size: 8 GiB speed: 2133 MT/s
  Device-3: CPU0-DIMM3 type: DDR4 size: 8 GiB speed: 2133 MT/s
  Device-4: CPU0-DIMM4 type: DDR4 size: 8 GiB speed: 2133 MT/s
  Device-5: CPU0-DIMM5 type: DDR4 size: 8 GiB speed: 2133 MT/s
  Device-6: CPU0-DIMM6 type: DDR4 size: 8 GiB speed: 2133 MT/s
  Device-7: CPU0-DIMM7 type: DDR4 size: 8 GiB speed: 2133 MT/s
  Device-8: CPU0-DIMM8 type: DDR4 size: 8 GiB speed: 2133 MT/s
CPU:
  Info: 12-core model: Intel Xeon E5-2690 v3 bits: 64 type: MT MCP cache:
    L2: 3 MiB
  Speed (MHz): avg: 1300 min/max: 1200/3500 cores: 1: 1198 2: 1902 3: 1198
    4: 1198 5: 1197 6: 1270 7: 1198 8: 1197 9: 1197 10: 1198 11: 1593 12: 1198
    13: 1199 14: 1198 15: 1200 16: 1199 17: 1199 18: 1885 19: 1198 20: 1198
    21: 1201 22: 1803 23: 1198 24: 1199

@malcolmlewis && @karlmistelberger:

So, in looking at the data, it looks like it took 13 minutes to completely run through the stress testing?

Reposting the question from before:

Code:
erlangen:~ # time echo “scale=5000; 4*a(1)” | bc -l

real 0m10.541s
user 0m10.539s
sys 0m0.001s
Is this command from Karl something that just tests up to “5000” and then shows the results?? I’m running out of time and it seems to be taking longer on the
18000 and above testing??

Output is the first 5000 digits of pi. 18000 digits:

**erlangen:~ #**  time echo "scale=18000; 4*a(1)" | bc -l  
3.141592653589793238462643383279502884197169399375105820974944592307\ 
81640628620899862803482534211706798214808651328230664709384460955058\ 
22317253594081284811174502841027019385211055596446229489549303819644\ 
28810975665933446128475648233786783165271201909145648566923460348610\
...
96174599705674507183320650345566440344904536275600112501843356073612\ 
22765949278393706478426456763388188075656121689605041611390390639601\ 
62022153684941092605387688714837989559999112099164 

real    4m35.511s 
user    4m35.495s 
sys     0m0.009s 
**erlangen:~ #**

bc uses 1 core.

Hi
My system did take about 13 minutes or so…

That is calculating PI to 5000 (scale) decimal places and timing how long it takes.


echo "scale=6; 4*a(1)" | bc -l

3.141592

OK, thanks for the replies. So this is another test phrase, one that does not have anything to do with running the “onemkl” or “prime95” apps?? It’s just a way to show relative speed of the cpu’s completion of a task . . . ??? But doesn’t do “analysis” or say that it “passes” ???

Hi
Correct, just different ways to see how the computer is running. The mprime (prime95) tool is very useful if overclocking cpu and/or ram to test it’s producing no errors, make small changes to speed, voltage etc and test.

@malcolmlewis:

Ah . . . OK . . . . Well, I’m definitely not “overclocking” this cpu. I did mess with that in my '00 PowerMac to get it to something like “1.2GHz” a few years back, trying to keep it “viable” in today’s internet . . . but then I guess the power supply took a dive . . . . So I can get mind-boggling PPC speeds for 20 minutes or so . . . then fade to black . . . .

I should be rotating back to TW tomorrow, so I’ll try to run the complete “onemkl” testing and see if it “passes” the entire test . . . got it up to “18K” as posted before and it “passed” up to that level before I . . . what, ctrl + z’d out of it . . . . Had other stuff to do. Life stuff.

Hi
Yes, that may be the limit for that cpu…

So if I use rust and multithreading, the bash example on this machine (only uses a single core) takes around 16 seconds, with rust and 24 cores it takes 1.12 seconds to 5000 decimal places :wink:

To quote a famous line . . . “rust never sleeps” . . . it just gets “rustier” . . . ???

**Leap-15-4:~ #** systemctl cat dup-15.4.service                      
**# /etc/systemd/system/dup-15.4.service**
[Unit] 
Description=Upgrade to Leap 15.4 

[Service] 
ExecStart=/usr/bin/zypper --releasever=15.4 dup --no-confirm --allow-vendor-change 
**Leap-15-4:~ #**
**Leap-15-4:~ #** journalctl -b -1 -u dup-15.4.service -g Pakete 
Jul 04 13:53:54 Leap-15-3 zypper[14482]: Installierte Pakete werden gelesen... 
Jul 04 13:53:55 Leap-15-3 zypper[14482]: Die folgenden 1380 Pakete werden aktualisiert: 
Jul 04 13:53:55 Leap-15-3 zypper[14482]: Die folgenden 60 Pakete werden durch eine ältere Version ausgetauscht: 
Jul 04 13:53:55 Leap-15-3 zypper[14482]: Die folgenden 5 Pakete werden die Architektur ändern: 
Jul 04 13:53:55 Leap-15-3 zypper[14482]: Die folgenden 35 Pakete werden den Anbieter ändern: 
Jul 04 13:53:55 Leap-15-3 zypper[14482]: Die folgenden 142 NEUEN Pakete werden installiert: 
Jul 04 13:53:55 Leap-15-3 zypper[14482]: Die folgenden 42 Pakete werden GELÖSCHT: 
Jul 04 13:53:55 Leap-15-3 zypper[14482]: 1380 Pakete werden aktualisiert, 60 werden zurückgestuft, 142 neue, 42 zu entfernen, 35  Anbieterwechsel, 5 Architekturwechsel. 
**Leap-15-4:~ #**

Installation start: Jul 04 14:07:47, finished: 14:15:55, systemd was reloaded 159 times.


**Leap-15-4:~ #** zypper lr -uE 
#  | Alias                 | Enabled | GPG Check | Priority | URI 
---+-----------------------+---------+-----------+----------+------------------------------------------------------------------------- 
 4 | packman               | Yes     | (r ) Yes  |   90     | https://ftp.fau.de/packman/suse/openSUSE_Leap_15.4/ 
 6 | repo-backports-update | Yes     | (r ) Yes  |   99     | https://mirrorcache-eu.opensuse.org/update/leap/15.4/backports/ 
11 | repo-non-oss          | Yes     | (r ) Yes  |   99     | https://mirrorcache-eu.opensuse.org/distribution/leap/15.4/repo/non-oss/ 
12 | repo-oss              | Yes     | (r ) Yes  |   99     | https://mirrorcache-eu.opensuse.org/distribution/leap/15.4/repo/oss/ 
14 | repo-sle-update       | Yes     | (r ) Yes  |   99     | https://mirrorcache-eu.opensuse.org/update/leap/15.4/sle/ 
16 | repo-update           | Yes     | (r ) Yes  |   99     | https://mirrorcache-eu.opensuse.org/update/leap/15.4/oss 
17 | repo-update-non-oss   | Yes     | (r ) Yes  |   99     | https://mirrorcache-eu.opensuse.org/update/leap/15.4/non-oss/ 
**Leap-15-4:~ #** 

@karlmistelberger, et al:

OK, 15.4 . . . been running the “alpha” or “beta” of that on my '09 MBPro for some time. NIce system. Not super high maintenance . . . if you know what I mean.

So, I tried again on running the “onemkl” cpu stress test on the Quad core Xeon Bloomfield processor . . . unlike @malcolmlewis’ run in 13 minutes . . . this was where I stopped it at roughly 1.5 hours of run-time . . . . Too much juice to waste running the “small block” V-8 engine of processors. It’s still showing “pass” for each of the line items . . . just doesn’t seem to be make quick work of it? I think I posted the previous run data that went up to “18K”?? This is from there to when I cut it . . . other stuff to do.

Perhaps the “PI” quiz challenge for faster test results???

20000  20016  4      205.938    25.9017  3.418252e-10 2.927800e-02   pass
20000  20016  4      208.470    25.5871  3.418252e-10 2.927800e-02   pass
20000  20016  4      204.738    26.0534  3.418252e-10 2.927800e-02   pass
22000  22008  4      280.752    25.2879  4.641425e-10 3.296392e-02   pass
22000  22008  4      288.182    24.6359  4.641425e-10 3.296392e-02   pass
22000  22008  4      285.260    24.8883  4.641425e-10 3.296392e-02   pass
25000  25000  4      407.443    25.5690  5.801093e-10 3.201574e-02   pass
25000  25000  4      417.947    24.9264  5.801093e-10 3.201574e-02   pass
25000  25000  4      427.604    24.3635  5.801093e-10 3.201574e-02   pass
26000  26000  4      471.705    24.8432  6.332790e-10 3.236595e-02   pass
26000  26000  4      453.952    25.8148  6.332790e-10 3.236595e-02   pass
26000  26000  4      406.987    28.7938  6.332790e-10 3.236595e-02   pass
27000  27000  4      535.475    24.5080  6.298261e-10 2.985655e-02   pass
27000  27000  4      531.519    24.6905  6.298261e-10 2.985655e-02   pass
^Z
[1]+  Stopped                 ./runme_xeon64


CPU, CPU Mark, Linpack GFlops
Intel Xeon W3565, 3346, 28
Intel Core i7-6700K, 8976, 222
Intel Xeon E5-2690 v4, 20077, 403

https://www.cpubenchmark.net/compare/Intel-Xeon-W3565-vs-Intel-i7-6700K-vs-Intel-Xeon-E5-2690-v4/1270vs2565vs2780

I am quite happy with the 6700K. It currently uses a SATA Crucial CT250MX500SSD1, which results in a higher CPU load compared to the NVME Samsung 950 PRO and thus substantially higher real time of zypper dup.

@karlmistelberger:

Thanks for that comparo. Yes, the 6700K seems to hit the “sweet spot” on the cost/benefit ratio and cost to run, etc. I think for my needs the Quad core unit is fine. It seems like in my case with the “ripping through packages” or “shuffling through them” . . . the weak point is that TW is installed into an HDD, rather than what might be a Crucial SSD that I could move it into . . . on a “rainy” day.

The other “issue” is that I think there might be a problem getting the 6700 into the '12 Mac Pro. I believe there are only a few cpu’s that are “supported” by that machine, it was a short list, as I recall from memory of the OWC site’s options. So, I’m not quite ready to build a DIYer machine just to be able to run TW upgrades . . . but, perhaps taking the time to flip the TW “/” over to SSD and . . . perhaps a slightly faster Quad-core cpu at some point. Stuff has to “break” before I try to “fix it.”