[lustre-discuss] Ideas for undergraduate thesis topic on Lustre and parallel filesystems

[Santiago Freire - InCo]

Dear Andreas,

Thank you for your reply. The project is more research oriented, with an estimated duration of one year. Regarding the projects you sent me, "improve read mirror selection policy for many mirrors" in particular caught my attention, and I'll look into it more this week. I'll also check out other projects in Jira.
Many thanks!
Santiago

On Dec 29 2024, at 6:51 am, Andreas Dilger <adilger at ddn.com> wrote:
> Hello Santiago,
> what kind of project are you looking for? More research oriented, or an implementation project? How much time will you spend on the project?
>
> There are a number of possible "implementation" projects in Jira marked "medium" difficulty you could work on for a thesis project:
>
> - https://jira.whamcloud.com/browse/LU-15250 ("RPC replay signatures")
> - https://jira.whamcloud.com/browse/LU-17957 ("set immutable with chmod")
> - https://jira.whamcloud.com/browse/LU-17973 ("improve read mirror selection policy for many mirrors")
> - https://jira.whamcloud.com/browse/LU-7880 ("add performance stats to statfs")
>
> Depending on your development experience, these might each take several weeks to complete, and some of them have performance aspects that depend on how they are implemented and would benefit from testing different algorithms, etc. to make a good report at the end.
>
> Cheers, Andreas
>
> > On Dec 19, 2024, at 16:55, Santiago Freire - InCo <sfreire at fing.edu.uy> wrote:
> > 
> >
> > You don't often get email from sfreire at fing.edu.uy. Learn why this is important (https://aka.ms/LearnAboutSenderIdentification)
> >
> >
> >
> >
> >
> >
> > Dear all,
> >
> > I'm an undergraduate student, cluster administrator and teaching assistant from Uruguay, and I'm about to start my undergraduate thesis - a requirement for completing my CS engineering degree. Given my background in HPC, I am planning to do focus my thesis on parallel filesystems, particularly Lustre.
> > However, I've been struggling to identify a topic to work on. If you have any suggestions for problems or research topics related to Lustre for me to work on, I'd be more than grateful. All ideas are welcome, and I will be forever thankful :)
> > Thank you in advance and have an excellent end of year.
> > Santiago
> > _______________________________________________
> > lustre-discuss mailing list
> > lustre-discuss at lists.lustre.org
> > http://lists.lustre.org/listinfo.cgi/lustre-discuss-lustre.org
> >
>
>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.lustre.org/pipermail/lustre-discuss-lustre.org/attachments/20241231/c977cb6d/attachment.htm>