Return-Path: nfsv4-bounces@ietf.org In-Reply-To: References: From: Benny Halevy Date: Fri, 14 Mar 2014 01:56:44 +0200 Message-ID: To: Michael Maymann , NFS list Cc: "nfsv4@ietf.org" Subject: Re: [nfsv4] PNFS questions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: multipart/mixed; boundary="===============7030815886579866100==" Errors-To: nfsv4-bounces@ietf.org Sender: "nfsv4" MIME-Version: 1.0 List-ID: --===============7030815886579866100== Content-Type: multipart/alternative; boundary=047d7bf0e158ce420a04f485b505 --047d7bf0e158ce420a04f485b505 Content-Type: text/plain; charset=ISO-8859-1 Michael, the IETF mailing list is really not the right place to ask about commercial solutions in general or linux solutions in particular. Cc'ing the linux-nfs mailing list. Comments in line below. On Thu, Mar 13, 2014 at 9:02 PM, Michael Maymann wrote: > Hi list, > > I'm setting up a new BIO-HPC facility and are looking for a fault > tolerant, scalable, fast and priceeffective storage solution. > > I have been looking at Coraid, Panasas and Lustre, but would like to > investigate my options also of building my own PNFS solution. Clients are > all Debian based. > > Can anyone recommend: > 1. If PNFS is ready for production or if I should stick with a commercial > vendor or build a Lustre solution ? > There is no pNFS server implementation in upstream linux yet. We maintain a development out-of-kernel tree in git:// linux-nfs.org/~bhalevy/linux-pnfs.git but it is not production ready. > 2. PNFS Metadata server SW prefferably for Debian ? > ditto #1 > 3. PNFS Diskshelfs to be used ? > In pNFS terminology, these are Data Servers. > 4. If AOE (ATA Over Ethernet) would make any sense together with PNFS ? > Maybe using the Block layout and AoE as the block storage transport protocol. > 5. PNFS GUI SW prefferably for Debian ? > ditto #1 > Thanks in advance :-) ! > > BR. > ~Maymann > > _______________________________________________ > nfsv4 mailing list > nfsv4@ietf.org > https://www.ietf.org/mailman/listinfo/nfsv4 > > --047d7bf0e158ce420a04f485b505 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable
Michael, the IETF mailing list is really not the r= ight place to ask about commercial solutions in general or linux solutions = in particular.
Cc'ing the linux-nfs mailing list. Comments in line= below.


On Thu, Mar 1= 3, 2014 at 9:02 PM, Michael Maymann <michael@maymann.org> = wrote:

Hi list,

I'm setting up a new BIO-HPC facility and are looking for a = fault tolerant, scalable, fast and priceeffective storage solution.

I have been looking at Coraid, Panasas and Lustre, but would= like to investigate my options also of building my own PNFS solution. Clie= nts are all Debian based.

Can anyone recommend:
1. If PNFS is ready for production or if I should stick with a commercial v= endor or build a Lustre solution ?


= There is no pNFS server implementation in upstream linux yet.
We maintain a development out-of-kernel tree in git://linux-nfs.org/~bhalevy/linux-pnfs.= git
but it is not production ready.
 <= /div>

2. PNFS Metadata server SW prefferably for Debian ?


ditto #1
 

3. PNFS Diskshelfs to be used ?


In pNFS termin= ology, these are Data Servers.
 

4. If AOE (ATA Over Ethernet) would make any sense together with PNFS ?
=


Maybe using the Block layout and AoE a= s the block storage transport protocol.
 

5. PNFS GUI SW prefferably for Debian ?


ditto #1

 

Thanks in advance :-) !

BR.
~Maymann


_______________________________________________
nfsv4 mailing list
nfsv4@ietf.org
h= ttps://www.ietf.org/mailman/listinfo/nfsv4


--047d7bf0e158ce420a04f485b505-- --===============7030815886579866100== Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Content-Disposition: inline _______________________________________________ nfsv4 mailing list nfsv4@ietf.org https://www.ietf.org/mailman/listinfo/nfsv4 --===============7030815886579866100==--