[Gate-users] writing output files

Wed Jan 6 17:58:41 CET 2021

Hi David, thank you for your input.
You were correct. I just tried running test simulations with and without ram disk – no change in real world time.

This is running on a Microsoft Azure VM, so I don’t know what type of disk storage they are actually using.

Josh

From: David Sarrut <David.Sarrut at creatis.insa-lyon.fr>
Sent: Wednesday, January 6, 2021 10:48 AM
To: Josh Knowland <jknowland at lucernodynamics.com>
Cc: Bryan McIntosh <mcintoshster at gmail.com>; gate-users at lists.opengatecollaboration.org
Subject: Re: [Gate-users] writing output files

Hi,

yes, Bryan is right, hits increase comp time and disk size, be sure you need it. There is now a "summary" output to print the number of hits, but that do not store the hits themselves. https://opengate.readthedocs.io/en/latest/data_output_management.html?highlight=summary#additional-output-summary

Some times ago, we tried to investigate the setOutFileSizeLimit, but it did not change anything. Nowadays, disks cache is really powerful, not fully sure that a ram disk will help (but I might be wrong). See here: https://github.com/OpenGATE/Gate/pull/339

David

On Wed, Jan 6, 2021 at 2:31 PM Josh Knowland <jknowland at lucernodynamics.com<mailto:jknowland at lucernodynamics.com>> wrote:
Hi Bryan – thanks a bunch, very good ideas.
I’m typically simulating gamma cameras, so singles are what I’m most interested in. I’ll try a ramdisk, though. That seems like it should speed up the sim.

-josh

From: Gate-users <gate-users-bounces at lists.opengatecollaboration.org<mailto:gate-users-bounces at lists.opengatecollaboration.org>> On Behalf Of Bryan McIntosh
Sent: Wednesday, January 6, 2021 8:26 AM
To: gate-users at lists.opengatecollaboration.org<mailto:gate-users at lists.opengatecollaboration.org>
Subject: Re: [Gate-users] writing output files

Hi Josh,

I did some benchmarking work in 2018 looking at the impact of the Spectre/Meltdown exploits on GATE simulation time; since those impacted I/O operations, I ran them on a RAM disk to remove my hard drive speed as a limiting factor. I've found that the following helps:

1) Remove hits or singles from the ROOT file output if you are simulating a PET system. In most simulations Hits aren't necessary, and if you are only using coincidences for creating projection data then you don't necessarily need the singles either. This also dramatically saves on disk space!

2) Since you have plenty of RAM, you can use a RAM disk as your output device. I used the guide at https://www.techrepublic.com/article/how-to-use-a-ramdisk-on-linux/, substituting 2 GB for whatever size of output data you needed. Once I created the RAM disk, I set my output path to it in the macro and everything worked seamlessly. This is a bit more dangerous since if the system crashes (unlikely) or if you lose power (depends on where you're located) you can lose some data, but it does offer a dramatic speed increase if you're running into storage speed limits.

I hope that this helps!

-Bryan McIntosh

On Wed, Jan 6, 2021 at 1:13 AM <gate-users-request at lists.opengatecollaboration.org<mailto:gate-users-request at lists.opengatecollaboration.org>> wrote:
Send Gate-users mailing list submissions to
        gate-users at lists.opengatecollaboration.org<mailto:gate-users at lists.opengatecollaboration.org>

To subscribe or unsubscribe via the World Wide Web, visit
        http://lists.opengatecollaboration.org/mailman/listinfo/gate-users
or, via email, send a message with subject or body 'help' to
        gate-users-request at lists.opengatecollaboration.org<mailto:gate-users-request at lists.opengatecollaboration.org>

You can reach the person managing the list at
        gate-users-owner at lists.opengatecollaboration.org<mailto:gate-users-owner at lists.opengatecollaboration.org>

When replying, please edit your Subject line so it is more specific
than "Re: Contents of Gate-users digest..."

Today's Topics:

   1. writing output files (Josh Knowland)
   2. Re: output unit (David Sarrut)
   3. Re: Gate settings for Linac photon beam simulation (David Sarrut)

----------------------------------------------------------------------

Message: 1
Date: Tue, 5 Jan 2021 22:05:11 +0000
From: Josh Knowland <jknowland at lucernodynamics.com<mailto:jknowland at lucernodynamics.com>>
To: "gate-users at lists.opengatecollaboration.org<mailto:gate-users at lists.opengatecollaboration.org>"
        <gate-users at lists.opengatecollaboration.org<mailto:gate-users at lists.opengatecollaboration.org>>
Subject: [Gate-users] writing output files
Message-ID:
        <BN8PR15MB3283B4A09BD77A39BFE2ACABAFD10 at BN8PR15MB3283.namprd15.prod.outlook.com<mailto:BN8PR15MB3283B4A09BD77A39BFE2ACABAFD10 at BN8PR15MB3283.namprd15.prod.outlook.com>>

Content-Type: text/plain; charset="utf-8"

All -
I have noticed that running simulations is non-linear in terms of real-world time vs simulated time. For instance, simulating 100 seconds takes longer than 100x a 1-second simulation.

My guess is that this could be due to file IO and the significantly larger file sizes resulting from longer simulations. I searched through the mailing list and the source code, but I couldn't find anything about the timing of file IO. I'd prefer not to use the "setOutFileSizeLimit" flag if I don't have to.

Since the computer I run GATE on has plenty of RAM, is it possible to increase the time between file IO writes in order to speed up the overall simulation?

-Josh

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.opengatecollaboration.org/pipermail/gate-users/attachments/20210105/da88531c/attachment-0001.html>

------------------------------
_______________________________________________
Gate-users mailing list
Gate-users at lists.opengatecollaboration.org<mailto:Gate-users at lists.opengatecollaboration.org>
http://lists.opengatecollaboration.org/mailman/listinfo/gate-users

--
David Sarrut, Phd
Directeur de recherche CNRS
CREATIS, UMR CNRS 5220, Inserm U1206
Centre de lutte contre le cancer Léon Bérard
28 rue Laënnec, 69373 Lyon cedex 08
Tel : 04 78 78 51 51 / 06 74 72 05 42
http://www.creatis.insa-lyon.fr/~dsarrut
_________________________________
 "2 + 2 = 5,  for extremely large values of 2"
_________________________________
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.opengatecollaboration.org/pipermail/gate-users/attachments/20210106/0d102c3b/attachment-0001.html>