Premysl Kouril
2018-Nov-12 11:51 UTC
[Gluster-users] Ceph or Gluster for implementing big NAS
Hi, We are planning to build NAS solution which will be primarily used via NFS and CIFS and workloads ranging from various archival application to more ?real-time processing?. The NAS will not be used as a block storage for virtual machines, so the access really will always be file oriented. We are considering primarily two designs and I?d like to kindly ask for any thoughts, views, insights, experiences. Both designs utilize ?distributed storage software at some level?. Both designs would be built from commodity servers and should scale as we grow. Both designs involve virtualization for instantiating "access virtual machines" which will be serving the NFS and CIFS protocol - so in this sense the access layer is decoupled from the data layer itself. First design is based on a distributed filesystem like Gluster or CephFS. We would deploy this software on those commodity servers and mount the resultant filesystem on the ?access virtual machines? and they would be serving the mounted filesystem via NFS/CIFS. Second design is based on distributed block storage using CEPH. So we would build distributed block storage on those commodity servers, and then, via virtualization (like OpenStack Cinder) we would allocate the block storage into the access VM. Inside the access VM we would deploy ZFS which would aggregate block storage into a single filesystem. And this filesystem would be served via NFS/CIFS from the very same VM. Any advices and insights highly appreciated Cheers, Prema -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.gluster.org/pipermail/gluster-users/attachments/20181112/26e384b1/attachment.html>
On 12/11/18 11:51, Premysl Kouril wrote:> > Hi, > > > We are planning to build NAS solution which will be primarily used via > NFS and CIFS and workloads ranging from various archival application > to more ?real-time processing?. The NAS will not be used as a block > storage for virtual machines, so the access really will always be file > oriented. > > > We are considering primarily two designs and I?d like to kindly ask > for any thoughts, views, insights, experiences. > > > Both designs utilize ?distributed storage software at some level?. > Both designs would be built from commodity servers and should scale as > we grow. Both designs involve virtualization for instantiating "access > virtual machines" which will be serving the NFS and CIFS protocol - so > in this sense the access layer is decoupled from the data layer itself. > > > First design is based on a distributed filesystem like Gluster or > CephFS. We would deploy this software on those commodity servers and > mount the resultant filesystem on the ?access virtual machines? and > they would be serving the mounted filesystem via NFS/CIFS. > > > Second design is based on distributed block storage using CEPH. So we > would build distributed block storage on those commodity servers, and > then, via virtualization (like OpenStack Cinder) we would allocate the > block storage into the access VM. Inside the access VM we would deploy > ZFS which would aggregate block storage into a single filesystem. And > this filesystem would be served via NFS/CIFS from the very same VM. > > > Any advices and insights highly appreciated > > > Cheers, > > Prema > >For just NAS, I'd suggest looking at some of the other Distributed File System projects such as MooseFS, LizardFS, BeeGFS (open source), weka.io, Exablox (proprietary), etc. They are perhaps more suited to a general purpose, unstructured NAS use with a mix of file sizes and workloads. GlusterFS would work but we found it only gave good enough performance on large files (>10MB) and was too slow with directories containing more that a thousand or so files. Cheers Alex -- This message is intended only for the addressee and may contain confidential information. Unless you are that person, you may not disclose its contents or use it in any way and are requested to delete the message along with any attachments and notify us immediately. This email is not intended to, nor should it be taken to, constitute advice. The information provided is correct to our knowledge & belief and must not be used as a substitute for obtaining tax, regulatory, investment, legal or any other appropriate advice. "Transact" is operated by Integrated Financial Arrangements Ltd. 29 Clement's Lane, London EC4N 7AE. Tel: (020) 7608 4900 Fax: (020) 7608 5300. (Registered office: as above; Registered in England and Wales under number: 3727592). Authorised and regulated by the Financial Conduct Authority (entered on the Financial Services Register; no. 190856). -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.gluster.org/pipermail/gluster-users/attachments/20181112/235a92f7/attachment.html>
Vlad Kopylov
2018-Nov-12 19:22 UTC
[Gluster-users] Ceph or Gluster for implementing big NAS
Good thing about gluster is that you have files as files. Whatever happens good old file access is still there - if you need backup, or rebuilding volumes - every replica brick has your files. As a contrary to object blue..something storage with separate metadata, if it gets lost/mixed you will be recovering it with magnifying glass... If you go with monster VM approach - hypervisor uses gfapi which is little faster then ceph on all simple tests. In really distributed environments ceph (multiple buildings or datacenters) read performance of ceph will kill the cluster. Ceph CPU and Mem consumption will surprise you against Gluster as well. For local FILE NAS (everything sitting in one room) something like BeeGFS LizardFS would be a best option. v On Mon, Nov 12, 2018 at 6:51 AM Premysl Kouril <premysl.kouril at gmail.com> wrote:> Hi, > > We are planning to build NAS solution which will be primarily used via NFS > and CIFS and workloads ranging from various archival application to more > ?real-time processing?. The NAS will not be used as a block storage for > virtual machines, so the access really will always be file oriented. > > We are considering primarily two designs and I?d like to kindly ask for > any thoughts, views, insights, experiences. > > Both designs utilize ?distributed storage software at some level?. Both > designs would be built from commodity servers and should scale as we grow. > Both designs involve virtualization for instantiating "access virtual > machines" which will be serving the NFS and CIFS protocol - so in this > sense the access layer is decoupled from the data layer itself. > > First design is based on a distributed filesystem like Gluster or CephFS. > We would deploy this software on those commodity servers and mount the > resultant filesystem on the ?access virtual machines? and they would be > serving the mounted filesystem via NFS/CIFS. > > Second design is based on distributed block storage using CEPH. So we > would build distributed block storage on those commodity servers, and then, > via virtualization (like OpenStack Cinder) we would allocate the block > storage into the access VM. Inside the access VM we would deploy ZFS which > would aggregate block storage into a single filesystem. And this filesystem > would be served via NFS/CIFS from the very same VM. > > Any advices and insights highly appreciated > > Cheers, > > Prema > _______________________________________________ > Gluster-users mailing list > Gluster-users at gluster.org > https://lists.gluster.org/mailman/listinfo/gluster-users-------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.gluster.org/pipermail/gluster-users/attachments/20181112/799eaa3a/attachment.html>