Posts by rebirther
log in
1) Message boards : News : Testing new apps started (Message 10043)
Posted 3 days ago by Profile rebirther
new cuda120 v39 TF mfaktc linux app is up

changelog:
- recompiled fix for sm_70
2) Message boards : Number crunching : Nvidia Tesla P100 Problems (Message 10040)
Posted 3 days ago by Profile rebirther
v39 with cc7 is up for linux
3) Message boards : Number crunching : Nvidia Tesla P100 Problems (Message 10038)
Posted 3 days ago by Profile rebirther


Yes, this was only testing the current plan_class with Cuda111 while there are some different things in config. It looks like the app was not compiled with cc7 but cc6. For cc10 we will have a separate app of course.


your current cuda120 app seems to support 5.0 (maybe, dont have any to test), 6.0, 6.1, 7.5, 8.6, 8.9, neglecting 7.0.

it seems clear that the project is only looking at the driver version or embedded cuda version for deciding which app to send. using an app_config WILL NOT WORK.


could you recompile the cuda120 app to include ALL CCs and add in 7.0?


Need to ask, windows and linux can be different.


I'm talking about the Linux cuda120 app


yes, compiling is in progress.
4) Message boards : Number crunching : Nvidia Tesla P100 Problems (Message 10035)
Posted 3 days ago by Profile rebirther


Yes, this was only testing the current plan_class with Cuda111 while there are some different things in config. It looks like the app was not compiled with cc7 but cc6. For cc10 we will have a separate app of course.


your current cuda120 app seems to support 5.0 (maybe, dont have any to test), 6.0, 6.1, 7.5, 8.6, 8.9, neglecting 7.0.

it seems clear that the project is only looking at the driver version or embedded cuda version for deciding which app to send. using an app_config WILL NOT WORK.


could you recompile the cuda120 app to include ALL CCs and add in 7.0?


Need to ask, windows and linux can be different.
5) Message boards : Number crunching : Nvidia Tesla P100 Problems (Message 10033)
Posted 3 days ago by Profile rebirther
all cuda120 tasks error out on my V100s. in the same way that crash's P100s were erroring.

did you omit support for CC 7.0 (titan V and V100 at this CC level)? as i pointed out before you need to explicitly add support for every CC level individually.


We can try to use cuda100 WUs with an app_config. Thats the oldest app which we are running.

<app_config> <app_version> <app_name>TF</app_name> <plan_class>cuda100</plan_class> <min_gpu_ram_mb>384</min_gpu_ram_mb> <gpu_ram_used_mb>384</gpu_ram_used_mb> <gpu_peak_flops_scale>0.22</gpu_peak_flops_scale> <cpu_frac>0.01</cpu_frac> </app_version> </app_config>


tried it. didnt work. this app config does not work. it will not limit the project to only getting the cuda100 app.

same results on a Titan V (same cc-7.0), instant error.

I'm guessing the problem is how the app was compiled without cc 7.0 support.


We can only try this

<app_config> <app_version> <app_name>TF</app_name> <plan_class>cuda111</plan_class> <min_gpu_ram_mb>384</min_gpu_ram_mb> <gpu_ram_used_mb>384</gpu_ram_used_mb> <gpu_peak_flops_scale>0.22</gpu_peak_flops_scale> <cpu_frac>0.01</cpu_frac> </app_version> </app_config>


Or the app was not compiled with cc7, P100 was working, right?


this method of using an app_config to try to force what app gets sent, do not work. i've never had this work reliably and it didnt work for crashtech when you suggested it last time. an app config is designed to enact some configuration on apps that are already downloaded. it does not impose any influence on what the project server actually sends me.

i think your app was not compiled with CC 7.0 support. and if you didn't do it on the latest app, I would guess you didnt do it on the older apps either.

P100 works as far as I know. but that's because you fixed it by recompiling the app to include the CC 6.0 support.

when compiling cuda apps you cannot just put a minimum value. you need to put ALL values that you want to support. to support everything from CC5.0+ you need to explicitly state every version (5.0, 5.2, 6.0, 6.1, 7.0, 7.5, 8.6, 8.9)

5.0 - GM100 Maxwell
5.2 - GM200 Maxwell
6.0 - GP100 Pascal
6.1 - GP102+ Pascal
7.0 - GV100 Volta (Titan V and V100)
7.5 - TU102+ Turing
8.0 - GA100 Ampere
8.6 - GA102+ Ampere
8.9 - AD102+ Ada Lovelace
9.0 - GH100 Hopper

and you're going to have to recompile again in the future to add 10.x something if you want to support the upcoming Blackwell GPUs.


Yes, this was only testing the current plan_class with Cuda111 while there are some different things in config. It looks like the app was not compiled with cc7 but cc6. For cc10 we will have a separate app of course.
6) Message boards : Number crunching : Nvidia Tesla P100 Problems (Message 10031)
Posted 3 days ago by Profile rebirther
all cuda120 tasks error out on my V100s. in the same way that crash's P100s were erroring.

did you omit support for CC 7.0 (titan V and V100 at this CC level)? as i pointed out before you need to explicitly add support for every CC level individually.


We can try to use cuda100 WUs with an app_config. Thats the oldest app which we are running.

<app_config> <app_version> <app_name>TF</app_name> <plan_class>cuda100</plan_class> <min_gpu_ram_mb>384</min_gpu_ram_mb> <gpu_ram_used_mb>384</gpu_ram_used_mb> <gpu_peak_flops_scale>0.22</gpu_peak_flops_scale> <cpu_frac>0.01</cpu_frac> </app_version> </app_config>


tried it. didnt work. this app config does not work. it will not limit the project to only getting the cuda100 app.

same results on a Titan V (same cc-7.0), instant error.

I'm guessing the problem is how the app was compiled without cc 7.0 support.


We can only try this

<app_config> <app_version> <app_name>TF</app_name> <plan_class>cuda111</plan_class> <min_gpu_ram_mb>384</min_gpu_ram_mb> <gpu_ram_used_mb>384</gpu_ram_used_mb> <gpu_peak_flops_scale>0.22</gpu_peak_flops_scale> <cpu_frac>0.01</cpu_frac> </app_version> </app_config>


Or the app was not compiled with cc7, P100 was working, right?
7) Message boards : Number crunching : Nvidia Tesla P100 Problems (Message 10029)
Posted 3 days ago by Profile rebirther
all cuda120 tasks error out on my V100s. in the same way that crash's P100s were erroring.

did you omit support for CC 7.0 (titan V and V100 at this CC level)? as i pointed out before you need to explicitly add support for every CC level individually.


We can try to use cuda100 WUs with an app_config. Thats the oldest app which we are running.

<app_config> <app_version> <app_name>TF</app_name> <plan_class>cuda100</plan_class> <min_gpu_ram_mb>384</min_gpu_ram_mb> <gpu_ram_used_mb>384</gpu_ram_used_mb> <gpu_peak_flops_scale>0.22</gpu_peak_flops_scale> <cpu_frac>0.01</cpu_frac> </app_version> </app_config>
8) Message boards : News : BOINC Games sprint 07/15/2024 08:00 (UTC) - 07/18/2024 08:00 (UTC) (Message 10027)
Posted 3 days ago by Profile rebirther
If you are a new member create an account over the website.
9) Message boards : Number crunching : Bases loaded (Message 10026)
Posted 5 days ago by Profile rebirther
S816
- n=500-600k
- runtime 5h07min-7h43min (AVX@3.8Ghz)
- 500-550k = 1800 credits
- 550-600k = 2300 credits
- Sierpinski / Riesel Base - long3
- deadline 4 days
- TOP5000 primes
10) Message boards : Number crunching : Bases loaded (Message 10025)
Posted 5 days ago by Profile rebirther
R842
- n=100-300k
- runtime 10min21-1h23min (AVX@3.8Ghz)
- 100-150k = 80 credits
- 150-200k = 150 credits
- 200-250k = 240 credits
- 250-300k = 300 credits
- Riesel Base
- deadline 3 days
11) Message boards : Number crunching : Bases loaded (Message 10024)
Posted 5 days ago by Profile rebirther
R275
- n=800k-1M
- runtime 5h57min-9h (AVX@3.8Ghz)
- 800-850k = 1950 credits
- 850-900k = 2100 credits
- 900-950k = 2400 credits
- 950k-1M = 2700 credits
- Sierpinski / Riesel Base - long2
- deadline 5 days
- TOP5000 primes
12) Message boards : Number crunching : couldn't get input files - download errors (Message 10023)
Posted 5 days ago by Profile rebirther
I have noticed this yeasterday in the WU generation.

414 WUs of S864 for long are affected and need to be reuploaded with an _a in wu name.
13) Message boards : Number crunching : Bases loaded (Message 10022)
Posted 6 days ago by Profile rebirther
R867
- n=500-600k
- runtime 3h37min-5h11min (AVX@3.8Ghz)
- 500-550k = 1000 credits
- 550-600k = 1250 credits
- Sierpinski / Riesel Base - average
- deadline 3 days
- TOP5000 primes
14) Message boards : Number crunching : Bases loaded (Message 10021)
Posted 6 days ago by Profile rebirther
S864
- n=500-600k
- runtime 4h40min-7h23min (AVX@3.8Ghz)
- 500-550k = 1800 credits
- 550-600k = 2200 credits
- Sierpinski / Riesel Base - long
- deadline 4 days
- TOP5000 primes
15) Message boards : Number crunching : Bases loaded (Message 10020)
Posted 6 days ago by Profile rebirther
S866
- n=500-600k
- runtime 3h49min-6h (AVX@3.8Ghz)
- 500-550k = 1500 credits
- 550-600k = 1800 credits
- Sierpinski / Riesel Base - long
- deadline 4 days
- TOP5000 primes
16) Message boards : Number crunching : Bases loaded (Message 10019)
Posted 6 days ago by Profile rebirther
S784
- n=400-600k
- runtime 1h52min-4h26min (AVX@3.8Ghz)
- 400-450k = 600 credits
- 450-500k = 700 credits
- 500-550k = 850 credits
- 550-600k = 1000 credits
- Sierpinski / Riesel Base - average2
- deadline 3 days
- TOP5000 primes
17) Message boards : Number crunching : Results of bases (Message 10018)
Posted 6 days ago by Profile rebirther
R91 tested to n=500k (400-500k)

3 primes found, 26 remain

43814*91^433332-1 found by [AF] Kalianthys (TOP5000)
223952*91^437353-1 found by Science United (TOP5000)
81030*91^440109-1 found by Aionel (TOP5000)
18) Message boards : Number crunching : Results of bases (Message 10017)
Posted 6 days ago by Profile rebirther
R75 tested to n=600k (500-600k)

2 primes found, 6 remain

968*75^522276-1 found by Science United (TOP5000)
856*75^530221-1 found by Cluster Physik (TOP5000)
19) Message boards : Number crunching : Bases loaded (Message 10015)
Posted 7 days ago by Profile rebirther
R60
- n=400-600k
- runtime 54min-2h27min (AVX@3.8Ghz)
- 400-450k = 300 credits
- 450-500k = 450 credits
- 500-550k = 500 credits
- 550-600k = 550 credits
- Sierpinski / Riesel Base - average3
- deadline 3 days
- TOP5000 primes
20) Message boards : News : Trial Factoring 75-76bit range started (Message 10014)
Posted 7 days ago by Profile rebirther
The next batch of 50k tests is uploading.

The runtime is now ~2h40min on a RX5500 XT and decreasing on higher ranges.


Next 20

Main page · Your account · Message boards


Copyright © 2014-2024 BOINC Confederation / rebirther