Attention: The Keycloak upgrade has been completed. As this was a major upgrade, there may be some unexpected issues occurring. Please report any issues you find to support by using the contact form found at https://www.ebrains.eu/contact/. Thank you for your patience and understanding. 


Changes for page Benchmarking

Last modified by paulapopa on 2022/10/04 08:19
From version 12.1
edited by paulapopa
on 2022/08/04 14:33
Change comment: There is no comment for this version
To version 1.1
edited by paulapopa
on 2022/08/03 15:01
Change comment: There is no comment for this version

Summary

Details

Page properties
Content
... ... @@ -1,5 +1,6 @@
1 1  == WIP ==
2 2  
3 +(% class="wikigeneratedid" %)
3 3  == TVB-INVERSION 1.0.0 ==
4 4  
5 5  === Sampling priors - Remote execution ===
... ... @@ -12,12 +12,11 @@
12 12  
13 13  These have been computed on DAINT-CSCS HPC, on a single node, with different configurations for the number of simulations and the number of workers.
14 14  
15 -(% border="2" cellspacing="10" style="margin-left:auto; margin-right:auto" %)
16 16  |=(% scope="row" %)(((
17 17  Model
18 18  )))|=(((
19 19  Sim length (s)
20 -)))|=Regions|=Nr simulations|=Nr workers|=Execution time (hh:mm)
20 +)))|=Nodes|=Nr simulations|=Nr workers|=Execution time (hh:mm)
21 21  |=MontbrioPazoRoxin|30|100|(((
22 22  30
23 23  )))|30|00:17
... ... @@ -28,33 +28,17 @@
28 28  |=MontbrioPazoRoxin|30|100|500|55|01:30
29 29  |=MontbrioPazoRoxin|30|100|600|55|01:45
30 30  |=MontbrioPazoRoxin|30|100|600|60|OOM
31 -|=MontbrioPazoRoxin|60|100|500|40|(((
32 -03:07
31 +|=MontbrioPazoRoxin|60|100|500|55|(((
32 +OOM
33 33  )))
34 -|(% scope="row" %)**MontbrioPazoRoxin**|60|100|500|55|OOM
35 -|= |= |= |= |= |=
36 -|= | | | | |
37 -|= | | | | |
38 -|= | | | | |
39 -|= | | | | |
40 -|= | | | | |
41 -|= | | | | |
42 -|= | | | | |
43 -|= | | | | |
44 -|= | | | | |
45 -|= | | | | |
46 -|= | | | | |
47 -|= | | | | |
48 -|= | | | | |
49 -|= | | | | |
50 50  
35 +==== ====
51 51  
52 -
53 53  ==== 2. Limitations ====
54 54  
55 55  * Reaching he memory limit on the CSCS node.
56 -** For a simulation of 30 seconds, we can fit 55 parallel workers in the available memory
57 -** For a simulation of 60 seconds, we can fit 40 parallel workers in the available memory
40 +** For a simulation of 30 seconds, we can fit 55 parallel workers in the available memory.
41 +** For a simulation of 60 seconds, we can fit 40 parallel workers in the available memory.
58 58  * Reaching the maximum connections on CSCS could happen during a run
59 59  ** connect once to launch the job
60 60  ** connect multiple times during the monitoring step to check the status of the job
Public

TVB Inversion