However, the bold or marketleading spot would go to the first to develop such even if it requires the development of a special fibre card and driver that even may be OS specific at first. Standards are set many times by whoever gets there first and grabs market share. For me it means a truly scalable, high performance file network on par and separate from session taffic networks. Just a lust I guess.
-----Original Message----- From: Dave Hitz [mailto:hitz@netapp.com] Sent: Monday, March 29, 1999 10:16 AM To: sschorr@homestead.com Cc: hitz@netapp.com; toasters@mathworks.com Subject: Re: NAC SAN
Why not a box with a special OS (ala WAFL) that runs on what servers recognize as a fibre hub, intercepts the SCSI calls from all attached servers, and does all the nifty stuff needed to make the SAN(?) look like locally attached storage? Advantage? Separate file I/O from TCP session traffic but still allow file sharing in some form or another.
SCSI is a raw disk block protocol. SCSI requests are very simple. They say things like "read block twelve-million and two" from disk, or "write block so-and-so to disk." There is no way in SCSI to express file level ideas like "create a file named foo" or "read the first block in a file name bar".
If you make the data look like locally attached storage, then all the different hosts will access it using their own local filesystems, so the different hosts will each use a different format, and the data on disk cannot be shared.
If you want the data to be shared, then you have to find some industry standard filesystem sharing techniques that everyone has implemented. Today the only such standards are the NAS standards (like NFS, HTTP, Windows Networking).
There are a variety of theoretically possible solutions that we could develop, possibly involving new file sharing protocols that could run over SCSI, or "global file systems" that everyone agrees to run on their system. But there are lots of tricky issues to resolve, and the standards work has barely begun, so even in the best possible case, it'd be many years before we could get a true heterogeneous data sharing solution.
Dave
For me it means a truly scalable, high performance file network on par and separate from session taffic networks. Just a lust I guess.
You can have that today -- just buy a Gigabit Ethernet switch and run only NFS and CIFS over it! Gigabit Ethernet runs at 100 MB/sec just like fibre channel does. Gigabit Ethernet is based on essentially the same high-speed switching fabric ASIC technology that fibre channel is.
Many of our high end customers do build separate machine room networks that the dedicate to high performance traffic between a farm of compute servers and a farm of NetApp filers.
It would help me think more about possible solutions if you could describe more about the problem you are trying to solve. If the problem is that file server traffic slows down other parts of your TCP/IP network, then using dedicated Gigabit switches is a great solution.
If you are trying to solve some other problems, then maybe it's not so good.
To me, SAN technology is ideally suited for things like attaching to lots and lots of disks, sharing tapes between multiple systems, so those are the benefits that we are hoping to achieve in our OEM relationship with Brocade.
Dave
On Mon, 29 Mar 1999, Dave Hitz wrote:
Gigabit Ethernet runs at 100 MB/sec just like fibre channel does.
125MB/s unless Gigabit != 1000 Mb in the Gigabit Ethernet context.
To me, SAN technology is ideally suited for things like attaching to lots and lots of disks, sharing tapes between multiple systems, so those are the benefits that we are hoping to achieve in our OEM relationship with Brocade.
Yes, but I'd still like to see only 1 protocol on the NAC. OK, 2 tops (NFS/CIFS replacement and maybe HTTP). Quite frankly I don't care whether the box thinks it is its local disk or a network disk (SAN vs. NAS).
Tom
-
Dave Hitz -
Sam Schorr -
tkaczma＠gryf.net