[lustre-discuss] Installing and Using Lustre in a Parallel Environment - For Parallel I/O

Dilger, Andreas andreas.dilger at intel.com
Wed Oct 12 23:46:11 PDT 2016

Probably the best brief overview of the Lustre architecture is available on the Wikipedia page:

For starting to use/test Lustre, there is a howto guide on the wiki:


Cheers, Andreas
Andreas Dilger
Lustre Principal Architect
Intel High Performance Data Division

On 2016/10/12, 06:27, "lustre-discuss on behalf of Aditya" <lustre-discuss-bounces at lists.lustre.org<mailto:lustre-discuss-bounces at lists.lustre.org> on behalf of adityasarma007 at gmail.com<mailto:adityasarma007 at gmail.com>> wrote:

Hello everyone!
I'm Aditya, a senior year undergraduate student working on parallel clustering algorithms. The professor I'm working with and I had come across various recent research papers such as this<http://dl.acm.org/citation.cfm?id=2807616> and this<http://dl.acm.org/citation.cfm?id=2834894> and we wish to incorporate a parallel IO in the distributed algorithm that we are working on. My professor and associated PhD scholars are not that familiar with parallel file systems in general, and lustre in specific, and neither am I.
I want to know the most efficient way to understand the basic working of lustre, set it up on my cluster and do a parallel reading of input. I have gone through the getting started guide but the user manual is too exhaustive for me. Any suggestions on how I should go about plugging in Lustre in my project?

[mage removed by sender.]
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.lustre.org/pipermail/lustre-discuss-lustre.org/attachments/20161013/2720aab5/attachment.htm>

More information about the lustre-discuss mailing list