From chair at gpfsug.org Fri Jan 17 14:26:32 2014 From: chair at gpfsug.org (Chair GPFS UG) Date: Fri, 17 Jan 2014 14:26:32 +0000 Subject: [gpfsug-discuss] GPFS User Group 2014 - Heads Up Message-ID: Hello all We've started to plan the next user group - currently slated for end of April / early May '14. We welcome open discussion regarding agenda items. We have our own list, though were not here to dictate. So if there's something that may be of interest to the group, please reply back and we'll do the leg work. Our aim is to make the User Group as relevant as possible for you. In addition the presentations from Super Computing 2013 have finally been signed off and are now available from the GPFS User Group Website under the 'Meetings/Presentations' Menu. Regards, Jez Tucker - Chair -------------- next part -------------- An HTML attachment was scrubbed... URL: From chair at gpfsug.org Fri Jan 17 14:39:35 2014 From: chair at gpfsug.org (Chair GPFS UG) Date: Fri, 17 Jan 2014 14:39:35 +0000 Subject: [gpfsug-discuss] Python API for GPFS Message-ID: Hi, Just putting feelers out as to the real-world desire for a Python API which wraps the GPFS C API? Jez -------------- next part -------------- An HTML attachment was scrubbed... URL: From orlando.richards at ed.ac.uk Fri Jan 17 14:41:58 2014 From: orlando.richards at ed.ac.uk (Orlando Richards) Date: Fri, 17 Jan 2014 14:41:58 +0000 Subject: [gpfsug-discuss] GPFS User Group 2014 - Heads Up In-Reply-To: References: Message-ID: <52D94136.2080301@ed.ac.uk> Hi Jez, Thanks for this! One topic I'd love to hear about is "authoritative gpfs tuning, tweaking and optimisation". Ideally, I'd like to hear from someone who has an authoritative knowledge of the various deep and dark GPFS tunables, who has used them in real world scenarios. A walk-through of case studies where they have tuned GPFS for performance (or other things), highlighting what they did, how they did it, and (critically) why they did it, would be absolutely lovely. Of course - I've no idea if such a person exists or not! -- Orlando On 17/01/14 14:26, Chair GPFS UG wrote: > Hello all > > We've started to plan the next user group - currently slated for end > of April / early May '14. > > We welcome open discussion regarding agenda items. > > We have our own list, though were not here to dictate. So if there's > something that may be of interest to the group, please reply back and > we'll do the leg work. > > Our aim is to make the User Group as relevant as possible for you. > > In addition the presentations from Super Computing 2013 have finally > been signed off and are now available from the GPFS User Group Website > under the 'Meetings/Presentations' Menu. > > Regards, > > Jez Tucker - Chair > > > _______________________________________________ > gpfsug-discuss mailing list > gpfsug-discuss at gpfsug.org > http://gpfsug.org/mailman/listinfo/gpfsug-discuss > -- -- Dr Orlando Richards Information Services IT Infrastructure Division Unix Section Tel: 0131 650 4994 skype: orlando.richards The University of Edinburgh is a charitable body, registered in Scotland, with registration number SC005336. From orlando.richards at ed.ac.uk Fri Jan 17 14:43:03 2014 From: orlando.richards at ed.ac.uk (Orlando Richards) Date: Fri, 17 Jan 2014 14:43:03 +0000 Subject: [gpfsug-discuss] Python API for GPFS In-Reply-To: References: Message-ID: <52D94177.7090901@ed.ac.uk> On 17/01/14 14:39, Chair GPFS UG wrote: > Hi, > > Just putting feelers out as to the real-world desire for a Python API > which wraps the GPFS C API? I'd buy that! :) -- -- Dr Orlando Richards Information Services IT Infrastructure Division Unix Section Tel: 0131 650 4994 skype: orlando.richards The University of Edinburgh is a charitable body, registered in Scotland, with registration number SC005336. From ckerner at ncsa.uiuc.edu Fri Jan 17 15:46:13 2014 From: ckerner at ncsa.uiuc.edu (Chad Kerner) Date: Fri, 17 Jan 2014 09:46:13 -0600 Subject: [gpfsug-discuss] Python API for GPFS In-Reply-To: References: Message-ID: <20140117154613.GB20390@logos.ncsa.illinois.edu> I would much rather write python than C any day. --Chad Chad Kerner, Systems Engineer Storage Enabling Technologies National Center for Supercomputing Applications University of Illinois Urbana-Champaign On Fri, Jan 17, 2014 at 02:39:35PM +0000, Chair GPFS UG wrote: > Hi, > > Just putting feelers out as to the real-world desire for a Python API which > wraps the GPFS C API? > > Jez > _______________________________________________ > gpfsug-discuss mailing list > gpfsug-discuss at gpfsug.org > http://gpfsug.org/mailman/listinfo/gpfsug-discuss From jonathan at buzzard.me.uk Fri Jan 17 15:57:22 2014 From: jonathan at buzzard.me.uk (Jonathan Buzzard) Date: Fri, 17 Jan 2014 15:57:22 +0000 Subject: [gpfsug-discuss] Python API for GPFS In-Reply-To: <20140117154613.GB20390@logos.ncsa.illinois.edu> References: <20140117154613.GB20390@logos.ncsa.illinois.edu> Message-ID: <1389974242.9787.9.camel@buzzard.phy.strath.ac.uk> On Fri, 2014-01-17 at 09:46 -0600, Chad Kerner wrote: > I would much rather write python than C any day. > That as it may be, for the sort of things I can think of wanting to use the GPFS API for, performance would be critical, and as such python would not be an appropriate solution. The two things I have in mind (and partly coded) are an mmsetfacl to provide a useful command line tool for setting ACL's and one that can be used on a directory recursively. The other is an mmfind but hooked into the inode scan feature of GPFS so ypu get the speed of the policy engine but in a simple command line for one off's. JAB. -- Jonathan A. Buzzard Email: jonathan (at) buzzard.me.uk Fife, United Kingdom. From chair at gpfsug.org Fri Jan 17 14:26:32 2014 From: chair at gpfsug.org (Chair GPFS UG) Date: Fri, 17 Jan 2014 14:26:32 +0000 Subject: [gpfsug-discuss] GPFS User Group 2014 - Heads Up Message-ID: Hello all We've started to plan the next user group - currently slated for end of April / early May '14. We welcome open discussion regarding agenda items. We have our own list, though were not here to dictate. So if there's something that may be of interest to the group, please reply back and we'll do the leg work. Our aim is to make the User Group as relevant as possible for you. In addition the presentations from Super Computing 2013 have finally been signed off and are now available from the GPFS User Group Website under the 'Meetings/Presentations' Menu. Regards, Jez Tucker - Chair -------------- next part -------------- An HTML attachment was scrubbed... URL: From chair at gpfsug.org Fri Jan 17 14:39:35 2014 From: chair at gpfsug.org (Chair GPFS UG) Date: Fri, 17 Jan 2014 14:39:35 +0000 Subject: [gpfsug-discuss] Python API for GPFS Message-ID: Hi, Just putting feelers out as to the real-world desire for a Python API which wraps the GPFS C API? Jez -------------- next part -------------- An HTML attachment was scrubbed... URL: From orlando.richards at ed.ac.uk Fri Jan 17 14:41:58 2014 From: orlando.richards at ed.ac.uk (Orlando Richards) Date: Fri, 17 Jan 2014 14:41:58 +0000 Subject: [gpfsug-discuss] GPFS User Group 2014 - Heads Up In-Reply-To: References: Message-ID: <52D94136.2080301@ed.ac.uk> Hi Jez, Thanks for this! One topic I'd love to hear about is "authoritative gpfs tuning, tweaking and optimisation". Ideally, I'd like to hear from someone who has an authoritative knowledge of the various deep and dark GPFS tunables, who has used them in real world scenarios. A walk-through of case studies where they have tuned GPFS for performance (or other things), highlighting what they did, how they did it, and (critically) why they did it, would be absolutely lovely. Of course - I've no idea if such a person exists or not! -- Orlando On 17/01/14 14:26, Chair GPFS UG wrote: > Hello all > > We've started to plan the next user group - currently slated for end > of April / early May '14. > > We welcome open discussion regarding agenda items. > > We have our own list, though were not here to dictate. So if there's > something that may be of interest to the group, please reply back and > we'll do the leg work. > > Our aim is to make the User Group as relevant as possible for you. > > In addition the presentations from Super Computing 2013 have finally > been signed off and are now available from the GPFS User Group Website > under the 'Meetings/Presentations' Menu. > > Regards, > > Jez Tucker - Chair > > > _______________________________________________ > gpfsug-discuss mailing list > gpfsug-discuss at gpfsug.org > http://gpfsug.org/mailman/listinfo/gpfsug-discuss > -- -- Dr Orlando Richards Information Services IT Infrastructure Division Unix Section Tel: 0131 650 4994 skype: orlando.richards The University of Edinburgh is a charitable body, registered in Scotland, with registration number SC005336. From orlando.richards at ed.ac.uk Fri Jan 17 14:43:03 2014 From: orlando.richards at ed.ac.uk (Orlando Richards) Date: Fri, 17 Jan 2014 14:43:03 +0000 Subject: [gpfsug-discuss] Python API for GPFS In-Reply-To: References: Message-ID: <52D94177.7090901@ed.ac.uk> On 17/01/14 14:39, Chair GPFS UG wrote: > Hi, > > Just putting feelers out as to the real-world desire for a Python API > which wraps the GPFS C API? I'd buy that! :) -- -- Dr Orlando Richards Information Services IT Infrastructure Division Unix Section Tel: 0131 650 4994 skype: orlando.richards The University of Edinburgh is a charitable body, registered in Scotland, with registration number SC005336. From ckerner at ncsa.uiuc.edu Fri Jan 17 15:46:13 2014 From: ckerner at ncsa.uiuc.edu (Chad Kerner) Date: Fri, 17 Jan 2014 09:46:13 -0600 Subject: [gpfsug-discuss] Python API for GPFS In-Reply-To: References: Message-ID: <20140117154613.GB20390@logos.ncsa.illinois.edu> I would much rather write python than C any day. --Chad Chad Kerner, Systems Engineer Storage Enabling Technologies National Center for Supercomputing Applications University of Illinois Urbana-Champaign On Fri, Jan 17, 2014 at 02:39:35PM +0000, Chair GPFS UG wrote: > Hi, > > Just putting feelers out as to the real-world desire for a Python API which > wraps the GPFS C API? > > Jez > _______________________________________________ > gpfsug-discuss mailing list > gpfsug-discuss at gpfsug.org > http://gpfsug.org/mailman/listinfo/gpfsug-discuss From jonathan at buzzard.me.uk Fri Jan 17 15:57:22 2014 From: jonathan at buzzard.me.uk (Jonathan Buzzard) Date: Fri, 17 Jan 2014 15:57:22 +0000 Subject: [gpfsug-discuss] Python API for GPFS In-Reply-To: <20140117154613.GB20390@logos.ncsa.illinois.edu> References: <20140117154613.GB20390@logos.ncsa.illinois.edu> Message-ID: <1389974242.9787.9.camel@buzzard.phy.strath.ac.uk> On Fri, 2014-01-17 at 09:46 -0600, Chad Kerner wrote: > I would much rather write python than C any day. > That as it may be, for the sort of things I can think of wanting to use the GPFS API for, performance would be critical, and as such python would not be an appropriate solution. The two things I have in mind (and partly coded) are an mmsetfacl to provide a useful command line tool for setting ACL's and one that can be used on a directory recursively. The other is an mmfind but hooked into the inode scan feature of GPFS so ypu get the speed of the policy engine but in a simple command line for one off's. JAB. -- Jonathan A. Buzzard Email: jonathan (at) buzzard.me.uk Fife, United Kingdom. From chair at gpfsug.org Fri Jan 17 14:26:32 2014 From: chair at gpfsug.org (Chair GPFS UG) Date: Fri, 17 Jan 2014 14:26:32 +0000 Subject: [gpfsug-discuss] GPFS User Group 2014 - Heads Up Message-ID: Hello all We've started to plan the next user group - currently slated for end of April / early May '14. We welcome open discussion regarding agenda items. We have our own list, though were not here to dictate. So if there's something that may be of interest to the group, please reply back and we'll do the leg work. Our aim is to make the User Group as relevant as possible for you. In addition the presentations from Super Computing 2013 have finally been signed off and are now available from the GPFS User Group Website under the 'Meetings/Presentations' Menu. Regards, Jez Tucker - Chair -------------- next part -------------- An HTML attachment was scrubbed... URL: From chair at gpfsug.org Fri Jan 17 14:39:35 2014 From: chair at gpfsug.org (Chair GPFS UG) Date: Fri, 17 Jan 2014 14:39:35 +0000 Subject: [gpfsug-discuss] Python API for GPFS Message-ID: Hi, Just putting feelers out as to the real-world desire for a Python API which wraps the GPFS C API? Jez -------------- next part -------------- An HTML attachment was scrubbed... URL: From orlando.richards at ed.ac.uk Fri Jan 17 14:41:58 2014 From: orlando.richards at ed.ac.uk (Orlando Richards) Date: Fri, 17 Jan 2014 14:41:58 +0000 Subject: [gpfsug-discuss] GPFS User Group 2014 - Heads Up In-Reply-To: References: Message-ID: <52D94136.2080301@ed.ac.uk> Hi Jez, Thanks for this! One topic I'd love to hear about is "authoritative gpfs tuning, tweaking and optimisation". Ideally, I'd like to hear from someone who has an authoritative knowledge of the various deep and dark GPFS tunables, who has used them in real world scenarios. A walk-through of case studies where they have tuned GPFS for performance (or other things), highlighting what they did, how they did it, and (critically) why they did it, would be absolutely lovely. Of course - I've no idea if such a person exists or not! -- Orlando On 17/01/14 14:26, Chair GPFS UG wrote: > Hello all > > We've started to plan the next user group - currently slated for end > of April / early May '14. > > We welcome open discussion regarding agenda items. > > We have our own list, though were not here to dictate. So if there's > something that may be of interest to the group, please reply back and > we'll do the leg work. > > Our aim is to make the User Group as relevant as possible for you. > > In addition the presentations from Super Computing 2013 have finally > been signed off and are now available from the GPFS User Group Website > under the 'Meetings/Presentations' Menu. > > Regards, > > Jez Tucker - Chair > > > _______________________________________________ > gpfsug-discuss mailing list > gpfsug-discuss at gpfsug.org > http://gpfsug.org/mailman/listinfo/gpfsug-discuss > -- -- Dr Orlando Richards Information Services IT Infrastructure Division Unix Section Tel: 0131 650 4994 skype: orlando.richards The University of Edinburgh is a charitable body, registered in Scotland, with registration number SC005336. From orlando.richards at ed.ac.uk Fri Jan 17 14:43:03 2014 From: orlando.richards at ed.ac.uk (Orlando Richards) Date: Fri, 17 Jan 2014 14:43:03 +0000 Subject: [gpfsug-discuss] Python API for GPFS In-Reply-To: References: Message-ID: <52D94177.7090901@ed.ac.uk> On 17/01/14 14:39, Chair GPFS UG wrote: > Hi, > > Just putting feelers out as to the real-world desire for a Python API > which wraps the GPFS C API? I'd buy that! :) -- -- Dr Orlando Richards Information Services IT Infrastructure Division Unix Section Tel: 0131 650 4994 skype: orlando.richards The University of Edinburgh is a charitable body, registered in Scotland, with registration number SC005336. From ckerner at ncsa.uiuc.edu Fri Jan 17 15:46:13 2014 From: ckerner at ncsa.uiuc.edu (Chad Kerner) Date: Fri, 17 Jan 2014 09:46:13 -0600 Subject: [gpfsug-discuss] Python API for GPFS In-Reply-To: References: Message-ID: <20140117154613.GB20390@logos.ncsa.illinois.edu> I would much rather write python than C any day. --Chad Chad Kerner, Systems Engineer Storage Enabling Technologies National Center for Supercomputing Applications University of Illinois Urbana-Champaign On Fri, Jan 17, 2014 at 02:39:35PM +0000, Chair GPFS UG wrote: > Hi, > > Just putting feelers out as to the real-world desire for a Python API which > wraps the GPFS C API? > > Jez > _______________________________________________ > gpfsug-discuss mailing list > gpfsug-discuss at gpfsug.org > http://gpfsug.org/mailman/listinfo/gpfsug-discuss From jonathan at buzzard.me.uk Fri Jan 17 15:57:22 2014 From: jonathan at buzzard.me.uk (Jonathan Buzzard) Date: Fri, 17 Jan 2014 15:57:22 +0000 Subject: [gpfsug-discuss] Python API for GPFS In-Reply-To: <20140117154613.GB20390@logos.ncsa.illinois.edu> References: <20140117154613.GB20390@logos.ncsa.illinois.edu> Message-ID: <1389974242.9787.9.camel@buzzard.phy.strath.ac.uk> On Fri, 2014-01-17 at 09:46 -0600, Chad Kerner wrote: > I would much rather write python than C any day. > That as it may be, for the sort of things I can think of wanting to use the GPFS API for, performance would be critical, and as such python would not be an appropriate solution. The two things I have in mind (and partly coded) are an mmsetfacl to provide a useful command line tool for setting ACL's and one that can be used on a directory recursively. The other is an mmfind but hooked into the inode scan feature of GPFS so ypu get the speed of the policy engine but in a simple command line for one off's. JAB. -- Jonathan A. Buzzard Email: jonathan (at) buzzard.me.uk Fife, United Kingdom. From chair at gpfsug.org Fri Jan 17 14:26:32 2014 From: chair at gpfsug.org (Chair GPFS UG) Date: Fri, 17 Jan 2014 14:26:32 +0000 Subject: [gpfsug-discuss] GPFS User Group 2014 - Heads Up Message-ID: Hello all We've started to plan the next user group - currently slated for end of April / early May '14. We welcome open discussion regarding agenda items. We have our own list, though were not here to dictate. So if there's something that may be of interest to the group, please reply back and we'll do the leg work. Our aim is to make the User Group as relevant as possible for you. In addition the presentations from Super Computing 2013 have finally been signed off and are now available from the GPFS User Group Website under the 'Meetings/Presentations' Menu. Regards, Jez Tucker - Chair -------------- next part -------------- An HTML attachment was scrubbed... URL: From chair at gpfsug.org Fri Jan 17 14:39:35 2014 From: chair at gpfsug.org (Chair GPFS UG) Date: Fri, 17 Jan 2014 14:39:35 +0000 Subject: [gpfsug-discuss] Python API for GPFS Message-ID: Hi, Just putting feelers out as to the real-world desire for a Python API which wraps the GPFS C API? Jez -------------- next part -------------- An HTML attachment was scrubbed... URL: From orlando.richards at ed.ac.uk Fri Jan 17 14:41:58 2014 From: orlando.richards at ed.ac.uk (Orlando Richards) Date: Fri, 17 Jan 2014 14:41:58 +0000 Subject: [gpfsug-discuss] GPFS User Group 2014 - Heads Up In-Reply-To: References: Message-ID: <52D94136.2080301@ed.ac.uk> Hi Jez, Thanks for this! One topic I'd love to hear about is "authoritative gpfs tuning, tweaking and optimisation". Ideally, I'd like to hear from someone who has an authoritative knowledge of the various deep and dark GPFS tunables, who has used them in real world scenarios. A walk-through of case studies where they have tuned GPFS for performance (or other things), highlighting what they did, how they did it, and (critically) why they did it, would be absolutely lovely. Of course - I've no idea if such a person exists or not! -- Orlando On 17/01/14 14:26, Chair GPFS UG wrote: > Hello all > > We've started to plan the next user group - currently slated for end > of April / early May '14. > > We welcome open discussion regarding agenda items. > > We have our own list, though were not here to dictate. So if there's > something that may be of interest to the group, please reply back and > we'll do the leg work. > > Our aim is to make the User Group as relevant as possible for you. > > In addition the presentations from Super Computing 2013 have finally > been signed off and are now available from the GPFS User Group Website > under the 'Meetings/Presentations' Menu. > > Regards, > > Jez Tucker - Chair > > > _______________________________________________ > gpfsug-discuss mailing list > gpfsug-discuss at gpfsug.org > http://gpfsug.org/mailman/listinfo/gpfsug-discuss > -- -- Dr Orlando Richards Information Services IT Infrastructure Division Unix Section Tel: 0131 650 4994 skype: orlando.richards The University of Edinburgh is a charitable body, registered in Scotland, with registration number SC005336. From orlando.richards at ed.ac.uk Fri Jan 17 14:43:03 2014 From: orlando.richards at ed.ac.uk (Orlando Richards) Date: Fri, 17 Jan 2014 14:43:03 +0000 Subject: [gpfsug-discuss] Python API for GPFS In-Reply-To: References: Message-ID: <52D94177.7090901@ed.ac.uk> On 17/01/14 14:39, Chair GPFS UG wrote: > Hi, > > Just putting feelers out as to the real-world desire for a Python API > which wraps the GPFS C API? I'd buy that! :) -- -- Dr Orlando Richards Information Services IT Infrastructure Division Unix Section Tel: 0131 650 4994 skype: orlando.richards The University of Edinburgh is a charitable body, registered in Scotland, with registration number SC005336. From ckerner at ncsa.uiuc.edu Fri Jan 17 15:46:13 2014 From: ckerner at ncsa.uiuc.edu (Chad Kerner) Date: Fri, 17 Jan 2014 09:46:13 -0600 Subject: [gpfsug-discuss] Python API for GPFS In-Reply-To: References: Message-ID: <20140117154613.GB20390@logos.ncsa.illinois.edu> I would much rather write python than C any day. --Chad Chad Kerner, Systems Engineer Storage Enabling Technologies National Center for Supercomputing Applications University of Illinois Urbana-Champaign On Fri, Jan 17, 2014 at 02:39:35PM +0000, Chair GPFS UG wrote: > Hi, > > Just putting feelers out as to the real-world desire for a Python API which > wraps the GPFS C API? > > Jez > _______________________________________________ > gpfsug-discuss mailing list > gpfsug-discuss at gpfsug.org > http://gpfsug.org/mailman/listinfo/gpfsug-discuss From jonathan at buzzard.me.uk Fri Jan 17 15:57:22 2014 From: jonathan at buzzard.me.uk (Jonathan Buzzard) Date: Fri, 17 Jan 2014 15:57:22 +0000 Subject: [gpfsug-discuss] Python API for GPFS In-Reply-To: <20140117154613.GB20390@logos.ncsa.illinois.edu> References: <20140117154613.GB20390@logos.ncsa.illinois.edu> Message-ID: <1389974242.9787.9.camel@buzzard.phy.strath.ac.uk> On Fri, 2014-01-17 at 09:46 -0600, Chad Kerner wrote: > I would much rather write python than C any day. > That as it may be, for the sort of things I can think of wanting to use the GPFS API for, performance would be critical, and as such python would not be an appropriate solution. The two things I have in mind (and partly coded) are an mmsetfacl to provide a useful command line tool for setting ACL's and one that can be used on a directory recursively. The other is an mmfind but hooked into the inode scan feature of GPFS so ypu get the speed of the policy engine but in a simple command line for one off's. JAB. -- Jonathan A. Buzzard Email: jonathan (at) buzzard.me.uk Fife, United Kingdom.