Hi Ben, Can you explain a bit more about what you mean by opposite of the post?
Regarding erasure coding, I'm confident in saying that not all erasure coding mechanisms were created equal, and I'm sure a deep dive on GPFS Native RAID could be provided for you that would clearly show how it is vastly superior in every way to the upcoming (1 year?) version of Swift erasure coding... On your other question regarding paging files out to an object store, that is actually the exact feature provided by the upcoming MCStore feature that was described at IBM Insight and IBM Enterprise in 2014. Unfortunately there isn't a lot of external info about these feature yet.... Dean Hildebrand IBM Almaden Research Center From: Ben Roeder <[email protected]> To: [email protected], gpfsug main discussion list <[email protected]> Date: 02/02/2015 08:19 AM Subject: Re: [gpfsug-discuss] First IBM developer blog post Sent by: [email protected] Hi, I would like to see the exact opposite of this post, esp with erasure coding coming to swift, instead of the 3 copies redundancy. I would like to see the ability to page files out to the object store, like you can to LTFS. as per http://www.zurich.ibm.com/sto/systems/bigdata.html but Openstack Swift instead of LTFS. Ben Roeder [email protected] -- ------------------------------------------------------------------------------- Sohonet Ltd web : http://www.sohonet.com/ The Surest Path support : [email protected] 24x7 : +44 20 7292 6909 +1 310 449 8610 +61 1800 77 5280 +64 9 951 8126 For full details of Sohonet's registered offices/addresses please visit http://www.sohonet.com/contact-us/ ------------------------------------------------------------------------------- > On 2 Feb 2015, at 16:02, Jez Tucker (Chair) <[email protected]> wrote: > > Hi all > > Our first IBM developer's blog post is up. > > http://www.gpfsug.org/gpfs-with-object/ > > Please take 10 secs to give some feedback via the mailing list. > Are we hitting the right level, spot, content etc.? > > What would you like to see? > > Very best, > > Jez > > > > > _______________________________________________ > gpfsug-discuss mailing list > gpfsug-discuss at gpfsug.org > http://gpfsug.org/mailman/listinfo/gpfsug-discuss _______________________________________________ gpfsug-discuss mailing list gpfsug-discuss at gpfsug.org http://gpfsug.org/mailman/listinfo/gpfsug-discuss
_______________________________________________ gpfsug-discuss mailing list gpfsug-discuss at gpfsug.org http://gpfsug.org/mailman/listinfo/gpfsug-discuss
