Main

Networks Archives

May 5, 2007

Small is Better as a SAN Best Practice

Nigel at the RupturedMonkey blog site wrote a post titled "Grrrrr big beefy manly SANs". He talks about a number of things that I think belong as part of a SAN best practices. The post has an interesting confessional tone where he discusses a certain fascination with large and complex networks - something that I think is common in our line of work. One of the things he discusses is limiting the number of ISLs in SANs as a way to reduce problems: I quote one of his paragraphs below:

"One thing that had a real influence on me was some work that I did for a telco company who have a policy of no ISLs in their SAN environments. A lot of people initially sniggle at the idea, and to be honest, I too raised an eyebrow when I first heard this. But I have to say that in their SAN environments problems were noticeably fewer and farther between, and when they did occur they were so much more limited in their scope and so much easier to troubleshoot."

A common mistake people make when thinking about SANs (especially iSCSI SANs) is assuming the same design principals and best practices apply to both LANs and SANs. A simple examination of the requirements indicates profound differences: For instance, with LANs, systems communicate frequently with other systems with the potential for any one system to communicate with every other system, but with SANs, systems usually don't have any reason to communicate with many other systems and storage. In fact, most SANs implement technology and processes to restrict potentially harmful communications. Put another way, LANs assume any-to-any communications whereas SANs assume a need-to-communicate-only operating model.

An excellent strategy for providing a need-to-communicate-only environment is to reduce the number of potential connections. In general, it is difficult to err in the pursuit of restricting SAN communications and this should be a design goal that is carried through the initial implementation and all subsequent changes to the SAN. Pursue policies that limit the number of ports in a SAN.

The use of ISLs should be limited as a way to prevent "connection creep" in the SAN. Core-edge designs with ISLs should not be blindly assumed as the best way to add nodes in the SAN. Instead think about using a single layer of switches that operate in parallel providing a single-hop environment. Stackable switches such as the Cisco 3750 for iSCSI or the QLogic 5200 for Fibre Channel have valuable scalability benefits.

Core-edge designs should be implemented with a goal of limiting the number of hops. The best core edge designs would have two hops between systems and storage. One way to do this is to connect storage ports to core switches and servers at the edge.

Consider using different SANs for disk I/O and backup. For instance, primary disk I/O might require dual paths with independent switches in those paths, whereas backup might only require a single path between servers and backup devices/systems. This limits the impact that backup and restore operations have on production disk I/O operations and solves many of the problems originating from the conflicting connectivity models for primary disk I/O and backup (the connectivity requirements for backup tend to be much broader than the connectivity requirements for disk access). For example, a dual-port HBA/NIC could be used to connect to the disk SAN while a single port HBA/NIC could be used to connect to the backup SAN.

February 21, 2008

Networking Webinar for iSCSI SANs

Lots of people have questions about the networking elements needed for iSCSI SANs. We are hosting a webinar on Tuesdsay Feb 26th at 12 noon Eastern to discuss network issues surrounding iSCSI.

Covered topics include:

  • Key criteria in selecting an appropriate network switch
  • Flow control, a definition and the benefits for iSCSI networks
  • Pros and cons of jumbo frames
  • Application of trunking and stacking of multiple switches
  • Connection of servers and arrays for maximum network fault tolerance

Click here to register.

March 17, 2008

Why did we agree to this anyway?

I've had a lot of people ask me why EqualLogic sold itself to Dell. Yes, the money mattered a lot, but it was more than the money. We didn't want to just take the money. If we were going to be sold, it needed to be someplace with a passion for what we were doing.

Matt Baker posted today on Direct2Dell about Dell's passion for iSCSI.

We are singing out of the same songbook - to quote Matt: So, when considering the economic benefits, the innovative storage models it facilitates, and the unrivaled flexibility, it is hard not to be hot on iSCSI

March 18, 2008

Farley Fireside Vlog: Keep SANs separate from LANs

Marc Farley talks about running an "air gap" network for your iSCSI SAN. (Thanks to Dave Siles for that analogy) . This advice holds for any iSCSI SAN, whether or not you are using Dell EqualLogic PS5000 series storage arrays or some other vendor's equipment.

April 3, 2008

Eric Schott on iSCSI/Fibre Channel coexistence

Questions about the future of both iSCSI and Fibre Channel. are popular with analysts and the press.

Eric Schott, Director of Product Management gives his opinions in the interview below, which was filmed by UberPulse. Many readers will likely be surprised by what he has to say, although to me it just demonstrates the kind of clear, realistic thinking he always has.

If you want to catch Eric in person, he will be presenting at SNW next Wednesday. Here is a link to a description of his presentation.

April 10, 2008

FCoE is a great dead end

My most recent post on Inside IT.

April 15, 2008

More on this FCoE thing

This blog was first posted here on Inside IT.

This FCoE thing is probably going to last for some time as a difference of opinions and perspectives. For those who wished I had kept my mouth shut (or keyboard locked), I was probably a little nastier a year ago with this post on my Storage @ Work blog.  Being a CREATURE OF HABIT,  I responded to last week's news from SNW with my usual open-minded and fair approach. 

Having said these snide things about FCoE, I am quite sure that it actually will be an excellent solution for lots of FC customers that need a migration path onto something less mortal than Fibre Channel.   The move to Ethernet-based SANs struck me as an obvious evolution a long time ago, after I heard the first FC bigot explain that FC was a channel and Ethernet was a network.  So I guess I shouldn't be too hard on FCoE, because it is a move in the right direction. Being an iSCSI technology bigot, it just seems like a unnecessary, cumbersome step, but to be fair, I tend to see the world through medium-sized business glasses.

I believe that most of the concerns people have had with iSCSI are based on the implementations that are available. They either don't exploit the iSCSI standard very well or they do not scale up to be a good fit for large-scale data processing.  I'm not buying the protocol arguments for FCoE. At the end of the day, I believe the brute force power of 10Gb Ethernet will be sufficient for iSCSI and that I would rather deal with the rare tuning problems that occur than the crushed-veggie-juice-on-papyrus methods of managing storage in an FC SAN.  I know Greg Ferro (see his comment on Dante's post)  agrees.

So where the heck do I stand today?  FCoE is as imperfect as FC, but it gets people headed where they need to be, which is Ethernet - and a lot better than FC.  There are going to be situations where taking a leap of faith to iSCSI is going to be a bit like riding a zipline.  You know its safe, but you don't want to attempt too much all at once.


May 8, 2008

Rock on, Dude!

DevCentral has an great post about the development of HA over the years. And he's a rocker too, which led me to a completely unrelated site that I liked.

About Networks

This page contains an archive of all entries posted to Storage @ Work in the Networks category. They are listed from oldest to newest.

NAS is the previous category.

Partners is the next category.

Many more can be found on the main index page or by looking through the archives.