This is Dmitri Kalintsev's Typepad Profile.
Join Typepad and start following Dmitri Kalintsev's activity
Join Now!
Already a member? Sign In
Dmitri Kalintsev
Recent Activity
Eric, I guess the "correct" answer is, as always, "it depends". In some environments it may be better to drop the defective frames and try to carry on, while in others it could make much better sense to cut off the offender as soon as possible. My "gut feel" preference is clearly on the side of isolating and fixing the root cause as soon as possible (because even if you carry on, you still know that you'll need an outage to fix the problematic hardware), but on the other hand if bad frames were dropped (meaning their impact is isolated to the directly affected system/s), and their volume wasn't very high, there would be more operational flexibility in regards to when to do the repairs in a more planned manner. As I said, "it depends". :) Regarding the idea of introducing an additional CRC for headers - I'm quite sceptical, as in addition to changes to networking standards it would almost certainly require changes to the networking hardware. Cheers, -- Dmitri
Hi Eric, I followed up with one of my vendor friends, who confirmed that their low latency data-only gear also operates in cut-through mode. His recommendation was that the Layer 1 errors are to be closely and pro-actively monitored, and identified bad talkers acted upon as soon as possible. Some network devices can trigger port actions (such as shut) based on error thresholds, based for e.g. on CRCs, but whether to use this or not would depend on the particular environment, of course. I agree with the recommendation to monitor and quick act on errors; and it should work equally well with NV. Cheers, -- Dmitri
Hi Eric, Cut-through switching is extremely uncommon in data networking, or at least it was until recently. Now that you mentioned it, I'm very interested in checking with the modern Ethernet "kings of low latency" to see if they are using cut-throgh. IMO, it is bad idea; I'm very much with Ivan on this one.
Hi Chuck, Thank you for the further details; the VMAX pedigree does indeed make things interesting. Reading http://www.emc.com/collateral/hardware/technical-documentation/h7113-vplex-architecture-deployment.pdf now; looks like I should be able to find answers to all further questions in there. :) -- Dmitri
Hi Chuck, As VPLEX is sitting in the data path, it surely has an opportunity to become a bottleneck for storage performance - both throughput and latency wise. How is this potential problem addressed by VPLEX's architecture? Cheers, -- Dmitri
Chuck, An excellent post. One note: we often go and do something because we've picked up on cues that are difficult to quantify, sometimes calling them "common sense". However, in many cases because they are difficult to quantify doesn't mean it's impossible. I believe there's at least one methodology out there that appears to works in many cases, enabling a structured opportunity discovery that yields results that can be reliably quantified - read: "it allows to figure out how much one can expect to make by addressing a particular discovered opportunity". It has been developed by Tony Ulwick, an HBS colleague and friend of the famous prof. C. Christensen, and been used to help companies across a wide variety of industries to discover (and justify, with hard, solid data) opportunities for growth. The theory behind the method is based on the notion of "jobs to be done", and the associated "desired outcomes". A person doing a job may not be able to provide you with a good idea on how to help him do it more efficiently, but they sure can tell you what steps they go through when doing the job, how important each step is, and how they know if they done well at each step. This information, when collected and analysed correctly, highlights the opportunities for improvement (important but underserved desired outcomes). Typically, the more important a particular underserved desired outcome is, the better chance that there's willingness to pay if you can help them do better. And because you can quite reliably judge if one or the other product improvement or a business initiative is likely to improve on satisfaction of an underserved important desired outcome, you can make business decisions with much more confidence. This isn't something that magically produces you an ROI, but it is something that can get you a long way to making a solid call on what the benefit is likely to be. Not only that, but it can help you find opportunities where you may have thought there were none. Don't want to appear soliciting or something, so no links. There is a book called "What customers want" that describes the methodology in detail, and there is a very good white paper on Tony's company web site, strategyn.com, titled "Outcome Driven Innovation", that summarises it. Cheers, -- Dmitri
Toggle Commented Jun 5, 2012 on Learning To Escape The ROI Trap at Chuck's Blog
Dmitri Kalintsev is now following The Typepad Team
Jun 4, 2012