Sure, if you put all routers in area 0 they will all get the updates, but you can also ameliorate an LSA storm by proper design of areas and stubbing those that don't go anywhere.
|
Does Verizon's Voyager stack up to the iPhone? |
|
|
5 IT skills that won't boost your salary 
[1,407]
Women 4 times more likely than men to cough up personal info [589]
Japan's 10 funniest tech-related commercials [Videos] [407]
Throwing away a promo CD is "unauthorized distribution"? [1,265]
Adults too quick to dismiss educational video games [682]
Attack of the iPhone clones [Slideshow] [578]
10 things IT needs to know about AJAX [1,258]
This Year's 25 Geekiest 25th Anniversaries [Slideshow] [409]
|
|
Algorithm Smalgorithm
This is moot if executives would simply dictate what the Internet is to be used for from a business perspective. In most cases, sans Web surfing, wild-eyed emailers and chat freaks, most organizations could probably do quite well with a 56 KB pipe. All these T-1s are merely helping fulfill an out-of-control user population's penchant for playing on the Internet.
efficient algorithm re. search engines
Schratboy speaks from the mindset of the 100% selfish!!
The Internet was originated by academics and curious web surfers. The search engines likewise by creative developers interested in expanding the horizons of creative endeavors.
The military got interested in it too.
When it became evident that the Internet potentially reduced costs and improved prospects for business interests, Internet became a favored venue for businesses as well.
Indeed, Internet features initiated marketplaces for innovative business -- with products in the Internet domain that would have been meaningless before the Internet existed.
Of course, the Internet also provided venues for selfish idiots like yourself to spew operatic, though hardly poetic, Odes to Intolerance and Hatred.
Here you degraded those who seek wider applications on this landscape, in the realms of communications, knowledge, exploration, excitement, and -- of all outrageous things conceivable -- fun!
The Internet also provides venues for defending the flourishing life and growth -- defense against the selfish narrow-mindedness of a Schratboy -- who would stunt the 'wild' non-business uses of this medium that gave it much of its vigor and generous availability.
Bravo to seekers, enablers, and progenerators of creative beauty and joy, who continue to expand the Internet's horizons, capacities, and availability.
Shame on you greedy self-centered -- on the stingy you who would condemn and deny the value of that pleasure and fulfillment.
Selfish and Context
My commentary was formulated from a business context. Personal Web surfing and Internet use on one's own time is perfectly fine.
I do think that you're a bit too enamored with the technology. The Internet is a poor substitute for living life and experiencing all of life's beauty. It may be a means to an end but is definitely not an end unto itself.
RE: Good Design
Exactly!
This is why we have flooding domains (areas) and Incremental SPF. What does this buy me that I don't already have within my current feature set? I would love for the creator of this algorithm to expand upon this.............
Routers in place for 10 years? Come on............
"Scalability may be limited by stuff you bought 10 years ago that you can't afford to replace yet"
A router in place for 10 years? I would argue that isn't the norm in this industry. So in some cases the link states need to be propagated end to end, so still not sure why this is so great (MPLS TE) and how it is better than Incremental SPF........
Cool
Sure you could do the same with proper design but that means talented individuals which means more money which meeans its too expensive. I know this will upset the cisco borg droids out there and Its likel cisco wouldnt adopt this anyway since it would mean not refreshing old networks. Got to keep the dollar rolling in for the big C.
RE:Cool
So you are advocating for dumb/cheap engineers with products which are so intelligent that a monkey could configure them? What happens when these products don't work as advertised?
Yes and No
The fundamental issue stated seems accurate to me. The use of the term 'flooding' and its ability to stimulate visions of whole communities swept away is overly histerical it seems to me. I would suggest that the strength of link-state algorithms such as Dykstra in OSPF may be refinable and the notion of storing only the least cost paths, note the plural, would be a way to reduce memory usage and route lookup. the problem is however that the least cost paths are only as good as their last update. It would seem the better solution would be to use static and truncted update methods such as stubby/not-so-stubby at the slow edges and let full link-state continue to run nearer the core. The current best and least cost paths are only as good as the full knowledge available from link updates. It is a catch-22 in that sense. But a refinement would be welcomed it seems to me.
Kudos - doesn't seem like much until you consider what it could
Combine XL with mesh networks, and unusual radio propogation and you have something interesting and suddenly usable. One problem with large (10's of thousands of nodes) mesh networks has always been how to contain the routing information, and where to draw "edge" boundaries. Just when you thought you'd had a clearly defined problems you have monkey wrenches tossed into your problem such as transient propogation conditions or mobile (possibly rapidly moving) nodes, unusual antenna patterns and other such problems. There was always a tradeoff between limiting connectivity and maximizing network efficiency. Consider the case of an airborn node flying through a mesh, both needing connectivity and providing it. Typical algorhythms would be unusably slow at converging and provide a route that is unusable by the time it is stable.
Will the vendors really buy it
Aim of vendors is to sell as many new boxes as quickly as possible. Will they really want to add it so that I can continue to run my old kit??
Post new comment