This is BRCDbreams's Typepad Profile.
Join Typepad and start following BRCDbreams's activity
Join Now!
Already a member? Sign In
BRCDbreams
Recent Activity
Perhaps the lesson is sweeping public policy change that requires systematic integration of disparate IT systems is an exercise in hubris. The reality maybe to think big but craft the implementation locally. If a policy vision can not be broken into small execution elements under local control, it is not a clear vision, but wishful thinking. Google, Amazon, et al do not have 30 year old infrastructure with legacy code for which the source does not exist. This is not uncommon in state government and even in the public sector. Integration at this scale must include a clear view of the complexity of the task, and adapt its implementation accordingly. I do not think this clarity was evident when the legislation was passed, let alone when the rules were promulgated by HHS.
Hi Eric, Thanks for posting this. I like the way you build this up from simple constructs. BTW, you mentioned for test 3: "Also note that test 3 had an interesting result. I believe this was the result of multiple flows being stacked on the same physical link while the other link remained under utilized. Perhaps this is one of those areas where SDN could help." I wanted to note that this is one good reason why the Brocade VCS Fabric has the Brocade ISL Trunk. The B-ISL Trunk automatically eliminates "hot spots" found in MLAG/LAG that are created by hashing techniques where an entire flow has to go on a physical link. Instead, B-ISL trunks uses frame stiping in hardware so all flows are uniformly distributed across all physical links. This means no hot spots and very high trunk utilization and its automatically configured so you don't have to program OpenFlow or an equivalent :-) Thanks for posting this content. It's very useful.
Chuck, There is an interesting post by Nicolas Carr on the noise vs signal aspects of data. http://www.roughtype.com/archives/2012/05/a_little_more_s.php The observation is more frequent measurements increase the noise level, not the signal. If data sets are big because they have more frequent measurements, you may find it a lot harder to understand the data than if the data samples are taken less frequently. Or, to continue with the "gravity" metaphor, if you collect too much data, your get so much gravity, you end up with a black hole from which nothing escapes, including information. In the end, businesses need to keep cost in mind. The fact you can sample data every second may increase the cost of actually understanding the data. Interesting ...
Toggle Commented May 31, 2012 on Does Data Have Gravity? at Chuck's Blog
Erik, Do I understand this correctly that the FCoE traffic is separated from the IP traffic on the ISL link? In short, this is NOT a converged network (FCP + IP over Ethernet), but a separate network physical path for FCoE and a second separate network physical path for IP traffic? Thanks.
Toggle Commented Nov 10, 2011 on FCoE Multihop and VE_Ports at Brass Tacks
Erik, And one other point I forgot to add to my earlier comment. I think my argument about FC routing preserving the air gap for the control plane meant that FC routing preserved the the air gap design goal of more than one independent control plan for storage traffic. IF you agree with that, AND, IF you further think your design for FCoE w/ vPC provides at least two independent control plans for FCoE traffic, then I'd say the title of your article would become "Emulating the SAN air gap for FCoE with vPC". Last, can vPC support more than 2 switches as we can with FC Routing? If so, that would provide good scalability.
Erik, First, full disclosuer, I work for Brocade. Thanks for showing this implementation of FCoE isolation. As I look at this, I think an argument can be made that in a lot of ways the isolation afforded is reminiscent of the isolation achived when using FC routing with multiple FC fabrics. With FC routing, each fabric control plane (and zoning and other policies) remains independent so Mr. Murphy-based configuration errors can't propogate beyond a single fabric. A good thing. For those reasons, FC routers, properly implemented, are seen as preserving the air gap architecture for fibre channel fabrics. I'd appreciate your thoughts on that. And, indeed simplifing configuration steps would be a virtue, so I'm interested in your experience with the Brocade VCS Ethernet fabric when you test that.
BRCDbreams is now following The Typepad Team
May 9, 2011