10Gb NAS vs Post Oriented SAN

[Steve Modica]

Here’s my take, from having watched the SAN schism occur back in the SGI days:
NAS systems had some issues. CPUS were slow. TCP took a lot of horse power and we just couldn’t build systems any larger to handle the NAS IO (think NFS).

The solution was to stop moving the data on the server, and just move the metadata. You can go get the data yourself. Now I have all my client cpus and IO busses working for me as well.

The problem of course is that it means 2 (or 3) networks and lots of complexity. Metadata (inodes and locking tokens) have to move around a TCP network, which can be wonky when clients die and leave things locked or the configuration manager loses control (zombie clients).

A NAS was always the most desirable, easy solution, it was just too hard.

Now, it’s not like that anymore. CPUS are fast, but more importantly, you have more of them than you (or any of the OS vendors) know how to use. They just sit there most of the time. So now, there are plenty of cores to run TCP and whatever other protocol stuff has to happen.

About the only legacy thing left over from the NAS days is the “direct attach” requirements some apps have. Some want specific locking calls. Some want inode access (protools 8 for example). Some have trouble writing extended attributes across NAS protocols (avid).

The great thing about a SAN is that it fools the apps and they think it’s direct attached, so all those things automatically go away.

All that being said, I think the SAN complexity will always be a burden. Ethernet is easy and it sucks in and adopts all the best elements of SAN. All your metadata stays in one place. Dead clients don’t kill the system and you only need one simple network.

What’s really blown me away in the last year has been the “re-rising” of iSCSI. The original iSCSI book (which I have a few copies of) predicates iSCSI on TCP offload cards. It clearly states that it’s expected that some special ASIC will be handling TCP. That didn’t catch on. A number of vendors doing that went away (alacritech, neteffect, neterion). iSCSI seemed doomed to fall behind things like FCoE (which skips the stack and uses an FC stack and a new form of flow control). However, now that we all have a gozillion cores on our laptops, iSCSI runs just fine and runs *really* fast. So we’re starting to see vendors push it again.
(we give people ready access to iSCSI with our Project Wrangler software).

Steve

Steve Modica
CTO, Small Tree Communications

[Neil Smith]

[Steve Modica] “All that being said, I think the SAN complexity will always be a burden. Ethernet is easy and it sucks in and adopts all the best elements of SAN. All your metadata stays in one place. Dead clients don’t kill the system and you only need one simple network. “

Nice summary of why and how things are changing, Steve … it’s funny how the wheel comes around full circle … I was in DEC during the 80’s when we started to implement VAX Clustering and DecNet and deal with emergence of TCP/IP LAN networking versus SNA/Token Ring … when I think back to the size of the machines and the costs involved to do what we now do on a couple of beefy Macs, PCs and Cat 6 10GbaseT networking, it’s kinda unreal.

It seems to me, that you don’t want to bet against Moore’s Law in the M&E industry now that digital acquisition and file-based workflow are the norm … 10Gb/s ethernet is pretty cost/effective now with 40 Gb/s on the horizon … IP over Thunderbolt 2 is also going to be a viable option as soon as certain company sorts out their SMB stack.

Interesting times indeed … are you going to the ‘Creative Storage Conference’ tomorrow? … if so, let’s grab a coffee some time.

Cheers,
Neil

Neil Smith
CEO
LumaForge LLC
Advanced Digital Workflow
323-850-3550
http://www.lumaforge.com

[Nat Jencks]

[Neil Smith] ” IP over Thunderbolt 2 is also going to be a viable option as soon as certain company sorts out their SMB stack.
“

Neil, I have been told that IP over thunderbolt will never be a good fast solution, since the Mac has to emulate an ethernet connection via software, and to do this stuff well at line speed requires purpose built hardware in the form of NICs… in other words, all those chips on the NIC are in fact doing something…

Do you feel that the current problems with IP over thunderbolt are just software issues and will be overcome if and when apple does a better job implementing this? I have never had good luck waiting for Apple to implement anything… even fixing critical bugs can take them years.

Curious if you think we will see fast reliable IP over thunderbolt in the near future.

best
-Nat

[Steve Modica]

I don’t think “never” is the right answer. I’m sure there will be something viable at some point. It’s just that right now, it doesn’t perform so well. I’m not sure if it’s framing activity or what. I haven’t profiled it.

Steve Modica
CTO, Small Tree Communications

[Gary Taylor]

Hi Steve,

I just read your 10GB and SMB3 blog post. Great stuff!

I am sure you have a lot going on but could you please profile this? Pretty please….

It would also be great to hear your take on this more recent thread.
https://forums.creativecow.net/readpost/197/860313

Thanks,
Gary

[Steve Modica]

Still working on this. The official samba code won’t release until Spring of next year.

Steve Modica
CTO, Small Tree Communications