Message boards : Theory Application : Windows App Efficiency
Message board moderation

To post messages, you must log in.

AuthorMessage
Profile Laurence
Project administrator
Project developer
Project tester
Avatar

Send message
Joined: 12 Sep 14
Posts: 1067
Credit: 329,449
RAC: 238
Message 6008 - Posted: 21 Feb 2019, 8:43:15 UTC

Now that the Windows VM restarts multiple times per day, I am interested to know the network usage on each boot and the boot time. Please post your experiences here.
ID: 6008 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile Magic Quantum Mechanic
Avatar

Send message
Joined: 8 Apr 15
Posts: 751
Credit: 11,610,376
RAC: 1,406
Message 6021 - Posted: 21 Feb 2019, 10:48:50 UTC - in response to Message 6008.  



It will still be a while before I can run these as you can see here.

AND since it is after 2am here I actually have the fastest ISP speed and had no problem doing anything else (such as getting over 40 LHC Theory tasks running jobs) and seeing my current speed .................

(almost 3am so I will check back later after I go pretend I am asleep)
ID: 6021 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Crystal Pellet
Volunteer tester

Send message
Joined: 13 Feb 15
Posts: 1182
Credit: 816,328
RAC: 288
Message 6055 - Posted: 22 Feb 2019, 15:30:58 UTC - in response to Message 6008.  

Now that the Windows VM restarts multiple times per day, I am interested to know the network usage on each boot and the boot time. Please post your experiences here.
I've no info about network usage during boot.
My impression is that the time between starting the VM and start processing the job is shorter as it was with the old (no container) VM.
Somewhere between 40 and 80 seconds.

Nothing to do with this new application, but with the vboxwrapper.
There is always 5 minutes lost after the result is copied to the shared folder.

2019-02-22 15:15:17 (6820): Guest Log: 14:12:20 2019-02-22: cranky: [INFO] Preparing output.

2019-02-22 15:15:17 (6820): VM Completion File Detected.
2019-02-22 15:15:17 (6820): Powering off VM.
2019-02-22 15:20:18 (6820): VM did not power off when requested.
2019-02-22 15:20:18 (6820): VM was successfully terminated.
2019-02-22 15:20:18 (6820): Deregistering VM. (boinc_0db2d5810825ed6c, slot#1)
2019-02-22 15:20:18 (6820): Removing network bandwidth throttle group from VM.
2019-02-22 15:20:18 (6820): Removing VM from VirtualBox.
15:20:24 (6820): called boinc_finish(0)

As you can see 'VM did not power off' is reported, but this is not true.
The VM is in power off state directly after 'Completion File Detected', so that 5 minutes waiting is for the cats.
ID: 6055 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile Magic Quantum Mechanic
Avatar

Send message
Joined: 8 Apr 15
Posts: 751
Credit: 11,610,376
RAC: 1,406
Message 6062 - Posted: 23 Feb 2019, 9:35:59 UTC - in response to Message 6055.  

Now that the Windows VM restarts multiple times per day, I am interested to know the network usage on each boot and the boot time. Please post your experiences here.
I've no info about network usage during boot.
My impression is that the time between starting the VM and start processing the job is shorter as it was with the old (no container) VM.
Somewhere between 40 and 80 seconds.

Nothing to do with this new application, but with the vboxwrapper.
There is always 5 minutes lost after the result is copied to the shared folder.


As you can see 'VM did not power off' is reported, but this is not true.
The VM is in power off state directly after 'Completion File Detected', so that 5 minutes waiting is for the cats.


That is not a task problem when running Theory tasks with X64 Windows OS CP
If you were running Theory tasks over at LHC you would know they always take those 5 minutes to go from
Powering off VM. to VM was successfully terminated. after that VM did not power off when requested.

But using a X86 Windows OS on those Theory tasks that does not happen and they only take one second to do that.

They have always been that way and I have ran 10's of thousands of those tasks.

.
ID: 6062 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote

Message boards : Theory Application : Windows App Efficiency


©2024 CERN