<html><body><p><font size="2">Rob,<br></font><font size="2"><br></font><font size="2">Talk to Jake Carroll from the University of Queensland, he has done a number of presentations at Scale User Groups of UQ’s MeDiCI data fabric which is based on Spectrum Scale and does very aggressive use of AFM.<br></font><font size="2"><br></font><font size="2">Their use of AFM is not only on campus, but to remote Storage clusters between 30km and 1500km away from their Home cluster. They have also tested AFM between Australia, Japan, and USA<br></font><font size="2"><br></font><font size="2">Sent from my iPhone<br></font><font size="2"><br></font><font size="2">> On 24 Nov 2020, at 01:20, Robert Horton <robert.horton@icr.ac.uk> wrote:<br></font><font size="2">> <br></font><font size="2">> Hi all,<br></font><font size="2">> <br></font><font size="2">> We're thinking about deploying AFM and would be interested in hearing<br></font><font size="2">> from anyone who has used it in anger - particularly independent writer.<br></font><font size="2">> <br></font><font size="2">> Our scenario is we have a relatively large but slow (mainly because it<br></font><font size="2">> is stretched over two sites with a 10G link) cluster for long/medium-<br></font><font size="2">> term storage and a smaller but faster cluster for scratch storage in<br></font><font size="2">> our HPC system. What we're thinking of doing is using some/all of the<br></font><font size="2">> scratch capacity as an IW cache of some/all of the main cluster, the<br></font><font size="2">> idea to reduce the need for people to manually move data between the<br></font><font size="2">> two.<br></font><font size="2">> <br></font><font size="2">> It seems to generally work as expected in a small test environment,<br></font><font size="2">> although we have a few concerns:<br></font><font size="2">> <br></font><font size="2">> - Quota management on the home cluster - we need a way of ensuring<br></font><font size="2">> people don't write data to the cache which can't be accomodated on<br></font><font size="2">> home. Probably not insurmountable but needs a bit of thought...<br></font><font size="2">> <br></font><font size="2">> - It seems inodes on the cache only get freed when they are deleted on<br></font><font size="2">> the cache cluster - not if they get deleted from the home cluster or<br></font><font size="2">> when the blocks are evicted from the cache. Does this become an issue<br></font><font size="2">> in time?<br></font><font size="2">> <br></font><font size="2">> If anyone has done anything similar I'd be interested to hear how you<br></font><font size="2">> got on. It would be intresting to know if you created a cache fileset<br></font><font size="2">> for each home fileset or just one for the whole lot, as well as any<br></font><font size="2">> other pearls of wisdom you may have to offer.<br></font><font size="2">> <br></font><font size="2">> Thanks!<br></font><font size="2">> Rob<br></font><font size="2">> <br></font><font size="2">> -- <br></font><font size="2">> Robert Horton | Research Data Storage Lead<br></font><font size="2">> The Institute of Cancer Research | 237 Fulham Road | London | SW3 6JB<br></font><font size="2">> T +44 (0)20 7153 5350 | E robert.horton@icr.ac.uk | W www.icr.ac.uk |<br></font><font size="2">> Twitter @ICR_London<br></font><font size="2">> Facebook: www.facebook.com/theinstituteofcancerresearch<br></font><font size="2">> <br></font><font size="2">> The Institute of Cancer Research: Royal Cancer Hospital, a charitable Company Limited by Guarantee, Registered in England under Company No. 534147 with its Registered Office at 123 Old Brompton Road, London SW7 3RP.<br></font><font size="2">> <br></font><font size="2">> This e-mail message is confidential and for use by the addressee only. If the message is received by anyone other than the addressee, please return the message to the sender by replying to it and then delete the message from your computer and network.<br></font><font size="2">> _______________________________________________<br></font><font size="2">> gpfsug-discuss mailing list<br></font><font size="2">> gpfsug-discuss at spectrumscale.org<br></font><font size="2">> <a href="http://gpfsug.org/mailman/listinfo/gpfsug-discuss">http://gpfsug.org/mailman/listinfo/gpfsug-discuss</a> <br></font><font size="2">> <br></font><BR>
</body></html>