Re: [Gluster-users] Hi new to Gluster
Hello Tarkeshwar, GlusterFS uses file level and range locks based on the operation. With the description you have given below, Gluster definitely seems to be the right choice for your workload. I recommend that you should try a proof of concept with Gluster under your application work load to experience the benefits. You can always reach out to GlusterFS community (this mailing list) for any queries and help regarding the same. Regards Nagaprasad. > On 15-Sep-2015, at 3:16 pm, M.Tarkeshwar Rao <tarkeshwa...@gmail.com> wrote: > > Hi Nagaprasad, > > Thanks for reply. > > > Nature of I/O workload your application is generating: > It is very high. Our product woking on files. It is collecting data(multi > process and multi threaded) from remote nodes and then processing it and then > it sending it to remote locations. > our Execution engine runs the processing busines logic. So huge no of > open,read,write,rename calls in the our code per second. > > recently we made it scalable as well, so our business logic runs horigontally > from multiple nodes. we are collecting it in common directory and reading it > from same directory for processing. > > We are using Veritas cluster file system. It is locking on directory level. > Since same directory accessed from multiple nodes. There is a delay in > processing. further it reducess performance drastically. > > For improving performance we made some changes in our application by breaking > the direcotories for collection and processing. > By this we got performance improvement. > > We feel if we change our file system we will get more improvement. Please > suggest. > > > Regards > Tarkeshwar > >> On Tue, Sep 15, 2015 at 2:27 PM, Nagaprasad Sathyanarayana >> <nsath...@redhat.com> wrote: >> Hello Tarakeshwar, >> >> Firstly, welcome to the Gluster community. >> >> Please visit >> http://www.gluster.org/community/documentation/index.php/GlusterFS_General_FAQ, >> which answers some of your queries about GlusterFS capabilities. >> If you could share with us the nature of I/O workload your application is >> generating, the performance need of your application, type of client access >> (NFS, CIFS etc.,) >> that users of your application need etc, we will be in a better position to >> guide. >> >> Regards >> Nagaprasad >> >> - Original Message - >> From: "M.Tarkeshwar Rao" <tarkeshwa...@gmail.com> >> To: gluster-users@gluster.org >> Sent: Tuesday, 15 September, 2015 12:15:23 PM >> Subject: [Gluster-users] Hi new to Gluster >> >> Hi all, >> We have a product which is written in c++ on Red hat. >> In production our customers using our product with Veritas cluster file >> system for HA and as sharded storage(EMC). >> Initially this product was run on only single node. In our last release we >> make it Scalable(more than one nodes). >> Due to excessive locking(CFS) we are not getting the performance. >> Can you please suggest Gluster will resolve our problem as it is distributed >> file system. >> is Gluster POSIX complined? >> Can we use it in Production? Pls suggest. >> If any other file system please suggest. >> Regards >> Tarkeshwar >> >> ___ >> Gluster-users mailing list >> Gluster-users@gluster.org >> http://www.gluster.org/mailman/listinfo/gluster-users > > ___ > Gluster-users mailing list > Gluster-users@gluster.org > http://www.gluster.org/mailman/listinfo/gluster-users ___ Gluster-users mailing list Gluster-users@gluster.org http://www.gluster.org/mailman/listinfo/gluster-users
Re: [Gluster-users] Hi new to Gluster
Hello Tarakeshwar, Firstly, welcome to the Gluster community. Please visit http://www.gluster.org/community/documentation/index.php/GlusterFS_General_FAQ, which answers some of your queries about GlusterFS capabilities. If you could share with us the nature of I/O workload your application is generating, the performance need of your application, type of client access (NFS, CIFS etc.,) that users of your application need etc, we will be in a better position to guide. Regards Nagaprasad - Original Message - From: "M.Tarkeshwar Rao"To: gluster-users@gluster.org Sent: Tuesday, 15 September, 2015 12:15:23 PM Subject: [Gluster-users] Hi new to Gluster Hi all, We have a product which is written in c++ on Red hat. In production our customers using our product with Veritas cluster file system for HA and as sharded storage(EMC). Initially this product was run on only single node. In our last release we make it Scalable(more than one nodes). Due to excessive locking(CFS) we are not getting the performance. Can you please suggest Gluster will resolve our problem as it is distributed file system. is Gluster POSIX complined? Can we use it in Production? Pls suggest. If any other file system please suggest. Regards Tarkeshwar ___ Gluster-users mailing list Gluster-users@gluster.org http://www.gluster.org/mailman/listinfo/gluster-users ___ Gluster-users mailing list Gluster-users@gluster.org http://www.gluster.org/mailman/listinfo/gluster-users
Re: [Gluster-users] [Gluster-devel] AFR arbiter volumes
Thanks Ravi for nicely explaining this. A question on the following section; "If 2 bricks are up and if one of them is the arbiter (i.e. the 3rd brick) and it blames the other up brick, then all FOPS will fail with ENOTCONN (Transport endpoint is not connected). If the arbiter doesn't blame the other brick, FOPS will be allowed to proceed. 'Blaming' here is w.r.t the values of AFR changelog extended attributes." Q: under what circumstances arbiter brick does/does not blame the other brick? Thanks Naga > On 09-Sep-2015, at 7:17 am, Ravishankar Nwrote: > > If 2 bricks are up and if one of them is the arbiter (i.e. the 3rd brick) and > it blames the other up brick, then all FOPS will fail with ENOTCONN > (Transport endpoint is not connected). If the arbiter doesn't blame the other > brick, FOPS will be allowed to proceed. 'Blaming' here is w.r.t the values of > AFR changelog extended attributes. ___ Gluster-users mailing list Gluster-users@gluster.org http://www.gluster.org/mailman/listinfo/gluster-users
Re: [Gluster-users] Command /etc/init.d/glusterd start failed
If you run # gluster volume info What is the value set for transport-type? Thanks Naga On 12-Apr-2014, at 7:33 am, 吴保川 wildpointe...@gmail.com wrote: Thanks, Joe. I found one of my machine has been assigned wrong IP address. This leads to the error. Originally, I thought the following error is critical: [2014-04-11 18:12:03.433371] E [rpc-transport.c:269:rpc_transport_load] 0-rpc-transport: /usr/local/lib/glusterfs/3.4.3/rpc-transport/rdma.so: cannot open shared object file: No such file or directory 2014-04-12 5:34 GMT+08:00 Joe Julian j...@julianfamily.org: On 04/11/2014 11:18 AM, 吴保川 wrote: [2014-04-11 18:12:05.165989] E [glusterd-store.c:2663:glusterd_resolve_all_bricks] 0-glusterd: resolve brick failed in restore I'm pretty sure that means that one of the bricks isn't resolved in your list of peers. ___ Gluster-users mailing list Gluster-users@gluster.org http://supercolony.gluster.org/mailman/listinfo/gluster-users ___ Gluster-users mailing list Gluster-users@gluster.org http://supercolony.gluster.org/mailman/listinfo/gluster-users