[Gluster-devel] Gluster-devel Digest, Vol 75, Issue 4

[Strahil Nikolov]

Hi Federico,

I'm not a dev, but based on my admin experience Gluster has  some weaknesses:

- Negative lookup is  bad ,  so if we got a fast way to identify a missing  file would be nice - but this  will be against the nature of Gluster and it's decentralised approach to metadata 
- Small  file performance boost is also a good one, as currently working with small files  is not as fast as some users  would like to. Checking the contents of a dir that is having 50000 small files is taking multiple times than the response of the bricks. For example  XFS responds  for  0.2 secs while FUSE needs at least 3 sec  (replica volume).

- Performance tuning is quite  hard, so  anything that could help the admin to set the optimal settings  would help alot.

Best Regards,
Strahil Nikolov 

На 11 юни 2020 г. 15:00:01 GMT+03:00, gluster-devel-request at gluster.org написа:
>Send Gluster-devel mailing list submissions to
>	gluster-devel at gluster.org
>
>To subscribe or unsubscribe via the World Wide Web, visit
>	https://lists.gluster.org/mailman/listinfo/gluster-devel
>or, via email, send a message with subject or body 'help' to
>	gluster-devel-request at gluster.org
>
>You can reach the person managing the list at
>	gluster-devel-owner at gluster.org
>
>When replying, please edit your Subject line so it is more specific
>than "Re: Contents of Gluster-devel digest..."
>
>
>Today's Topics:
>
>   1. Introducing me, questions on general improvements in gluster
>      re. latency and throughput (Federico Strati)
>
>
>----------------------------------------------------------------------
>
>Message: 1
>Date: Thu, 11 Jun 2020 09:51:52 +0200
>From: Federico Strati <strati.federico at gmail.com>
>To: gluster-devel at gluster.org
>Subject: [Gluster-devel] Introducing me, questions on general
>	improvements in gluster re. latency and throughput
>Message-ID: <0c283930-ebb7-4cc0-a063-2fb40a8b805e at gmail.com>
>Content-Type: text/plain; charset=utf-8; format=flowed
>
>Dear All,
>
>I just started working for a company named A3Cube, who produces HPC 
>supercomputers.
>
>I was assigned the task to investigate which improvements to gluster
>are 
>viable
>
>in order to lead to overall better performance in latency and
>throughput.
>
>I'm quite new to SDS and so pardon me if some questions are naive.
>
> From what I've understood so far, possible bottlenecks are
>
>in FUSE and transport.
>
>Generally speaking, if you have time to just drop me some pointers,
>
>1] FUSE + splice has never been considered (issue closed without real 
>discussions)
>
>(probably because it conflicts with the general architecture and in 
>particular
>
>with the write-behind translator)
>
>Recently, it has been announced a new userspace fs kernel module, ZUFS,
>
>whose aim
>
>is to zero copy and improving vastly over FUSE: would you be interested
>
>in investigating it ?
>
>(ZUFS: https://github.com/NetApp/zufs-zuf ; 
>https://lwn.net/Articles/756625/)
>
>2] Transport over RDMA (Infiniband) has been recently dropped:
>
>may I ask you what considerations have been made ?
>
>3] I would love to hear what you consider real bottlenecks in gluster
>
>right now regarding latency and thruput.
>
>Thanks in advance
>
>Kind regards
>
>Federico
>
>
>
>------------------------------
>
>_______________________________________________
>Gluster-devel mailing list
>Gluster-devel at gluster.org
>https://lists.gluster.org/mailman/listinfo/gluster-devel
>
>End of Gluster-devel Digest, Vol 75, Issue 4
>********************************************