Hello Jonathan, > Does that mean the mm-command layer > is going away? Will I in the future going to forced to use some "naff" > GUI layer to administer a GPFS cluster? Frankly I am quite happy using > the mm-command layer thank you very much and would like to keep it that > way and just be able to ignore the GUI. I do appreciate I might be > somewhat old school in that view but never the less I view GUI > administration of things with disdain.
The Native Rest API is a new feature that is meant to replace mm-commands in the long-term. The Native Rest API feature is being delivered in phases. For the 5.2.2.0 GA, not all the functionality that mm-commands expose will be available in the Native Rest API. Due to this limitation, all mm-commands will remain available, and complete co-existence is supported by the Native Rest API with mm-commands (meaning you can run both at the same time). The Native Rest API will not require GUI administration. With the Native Rest API, there is a new CLI that has a similar look to its equivalent mm-command (-N options, -F options with stanza files, and many of the same flags per command, just a different invocation). > Secondly at the moment the Rest API requires installing the GUI. Does > the "native" bit of the title mean that requirement is going away and > there will be a Rest API without the need for the additional complexity > of the GUI nodes? Or is the mm-command layer going away and yes you will > need the extra complexity of the GUI because you are going to have to > suck up administering the system with a GUI? The Native Rest API will not require installing the GUI. It does require a new RPM+service, but it does not expose a GUI itself. Once the Native Rest API exposes the functionality that the GUI requires, the GUI itself will rely on the Native Rest API (assuming you want a GUI running), as well as other internal components that today rely on mm-commands. Thanks, Luis Teran IBM Storage Scale Development From: gpfsug-discuss <[email protected]> on behalf of [email protected] <[email protected]> Date: Tuesday, August 20, 2024 at 9:10 AM To: [email protected] <[email protected]> Subject: [EXTERNAL] gpfsug-discuss Digest, Vol 149, Issue 13 Send gpfsug-discuss mailing list submissions to [email protected] To subscribe or unsubscribe via the World Wide Web, visit http://gpfsug.org/mailman/listinfo/gpfsug-discuss_gpfsug.org or, via email, send a message with subject or body 'help' to [email protected] You can reach the person managing the list at [email protected] When replying, please edit your Subject line so it is more specific than "Re: Contents of gpfsug-discuss digest..." Today's Topics: 1. Native Rest API (Jonathan Buzzard) 2. Re: Announcement: Scale 5.2.1 is out! (Amy Hirst) ---------------------------------------------------------------------- Message: 1 Date: Tue, 20 Aug 2024 15:36:41 +0100 From: Jonathan Buzzard <[email protected]> To: gpfsug main discussion list <[email protected]> Subject: [gpfsug-discuss] Native Rest API Message-ID: <[email protected]> Content-Type: text/plain; charset=UTF-8; format=flowed I just had an email from IBM about technology preview of the "Native Rest API" feature in 5.2.1.0 There are at least two interrelated and important questions that are not answered in the web page about this "Native Rest API" feature IMHO. Firstly the page says it "eliminates" the need to administer the Scale cluster with the mm-command layer. Does that mean the mm-command layer is going away? Will I in the future going to forced to use some "naff" GUI layer to administer a GPFS cluster? Frankly I am quite happy using the mm-command layer thank you very much and would like to keep it that way and just be able to ignore the GUI. I do appreciate I might be somewhat old school in that view but never the less I view GUI administration of things with disdain. Secondly at the moment the Rest API requires installing the GUI. Does the "native" bit of the title mean that requirement is going away and there will be a Rest API without the need for the additional complexity of the GUI nodes? Or is the mm-command layer going away and yes you will need the extra complexity of the GUI because you are going to have to suck up administering the system with a GUI? JAB. -- Jonathan A. Buzzard Tel: +44141-5483420 HPC System Administrator, ARCHIE-WeSt. University of Strathclyde, John Anderson Building, Glasgow. G4 0NG ------------------------------ Message: 2 Date: Tue, 20 Aug 2024 16:06:36 +0000 From: Amy Hirst <[email protected]> To: gpfsug main discussion list <[email protected]>, CHRIS MAESTAS <[email protected]> Subject: Re: [gpfsug-discuss] Announcement: Scale 5.2.1 is out! Message-ID: <sn6pr15mb24950e1b9655291e454a062183...@sn6pr15mb2495.namprd15.prod.outlook.com> Content-Type: text/plain; charset="utf-8" I?m always happy to see these emails. Well done, team! Thank you, Amy (Purdy) Hirst Vice President IBM Storage Software, Site Reliability Engineering, and User Experience She/Her/Hers Assistant: Michelle Garcia Diaz ([email protected]<mailto:[email protected]>) IBM -- From: gpfsug-discuss <[email protected]> on behalf of CHRIS MAESTAS <[email protected]> Date: Monday, August 19, 2024 at 7:49?PM To: [email protected] <[email protected]> Subject: [EXTERNAL] [gpfsug-discuss] Announcement: Scale 5.2.1 is out! For a summary of changes see here! If you miss: dstat ?gpfs ?gpfs-ops Maybe your new favorite command is mmpstat! And if you want those expelled nodes to stay down, now they will! That is unless you run: mmexpelnode -r/?reset The new and improved For a summary of changes see here<https://www.ibm.com/docs/en/storage-scale/5.2.1?topic=summary-changes>! If you miss: dstat ?gpfs ?gpfs-ops Maybe your new favorite command is mmpstat<https://www.ibm.com/docs/en/storage-scale/5.2.1?topic=reference-mmpstat-command>! And if you want those expelled nodes to stay down, now they will! That is unless you run: mmexpelnode -r/?reset The new and improved Cluster Export Services (CES) S3 is here<https://community.ibm.com/community/user/storage/blogs/mike-kieran/2024/08/16/ibm-storage-scale-software-delivers-next-generatio?CommunityKey=1142f81e-95e4-4381-95d0-7977f20d53fa>! Think of it as High Performance Object (HPO) 2.0 now also running on VMs and bare metal! Check out the test measurements here<https://community.ibm.com/community/user/storage/blogs/rogelio-rivera-gutierrez/2024/04/25/ibm-storage-scale-performance-ces-s3-tech-preview> where you can get 60 GB/s of read performance. Yes, that?s a byte which is a lot of bits! If you want to get Scale storage services on arm64 platforms now, you can! Unofficially, let?s race to running it on your Pi! Scale training has also been updated and it?s available on IBM training and Coursera! Check out this<https://community.ibm.com/community/user/storage/blogs/james-collier/2024/08/15/elevate-your-data-management-skills-with-new-ibm-s> blog post for more information on classes and subscription options! -- The Chief Troublemaker 8) -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://gpfsug.org/pipermail/gpfsug-discuss_gpfsug.org/attachments/20240820/12ccbb9e/attachment.htm > ------------------------------ Subject: Digest Footer _______________________________________________ gpfsug-discuss mailing list gpfsug-discuss at gpfsug.org http://gpfsug.org/mailman/listinfo/gpfsug-discuss_gpfsug.org ------------------------------ End of gpfsug-discuss Digest, Vol 149, Issue 13 ***********************************************
_______________________________________________ gpfsug-discuss mailing list gpfsug-discuss at gpfsug.org http://gpfsug.org/mailman/listinfo/gpfsug-discuss_gpfsug.org
