Combinatorics Difficulties (redux)

Thanks to Don's comment, I was able to continue investigating this interesting problem.


Combinatorics Difficulties

Over the weekend I was thinking of how to generalize the number of steps needed to isolate a physical issue in a network. My reasoning was thus: if device A appears to be failing, device B is a known-working device, and cable sets {a1, a2, a3} and {b1, b2, b3} are connected to A and B, respectively, then isolating the problem will require swapping all like cables between the two devices to determine under which circumstances the failure persists. I also assumed that a thorough investigation would require all possible combinations be explored.

Einstein/Zebra Puzzle

Last night I stumbled upon a well-known, inferential logic puzzle called the Zebra Puzzle or Einstein's Puzzle.

According to my GoogleDocs Spreadsheet revision history I solved this puzzle in about 20 minutes, making it a nice armchair diversion.



Jonny and the Cell (pt. 2.5)

Much has happened since my fledgling foray into the world of wireless networking. After finding and confirming facts for over a month I set to work developing a method that could be used to assess the quality of existing networks yet still provide a firm basis for the design of new ones. I derived a set of equations that could be used to answer the most difficult questions I had about access point placement and then developed a tool to simplify the solutions. I began using a grid, optimized for multi-floor, multi-unit deployments, to predict where access points would need to be placed and provide documentation that field technicians could follow during deployment. And after all this I am still developing my method and still finding evidence of its predictive power.


Confirming BGP Advertisement

Needing a quick and easy way of confirming routing for a given prefix, I decided to write an Expect script that would check all of the major route-servers with telnet capability for BGP best path information and dampening/penalties. This is really useful for documenting evidence of a new route's completion.

Documenting Layer-3 Topologies

I recently decided to try making a set of scripts to generate a layer-3 diagram given a list of routers. My method was simple: obtain a list of connected routes and interfaces for each router in a list, find those networks that are shared by two or more routers, then spit this information out in a format that graphviz can use and generate a diagram. My goal was to make this work on a Metro Area Network of OSPF-speaking routers.


Svilnog aagrnams

Last Saturday I was sitting at Monical's Pizza, scarfing down a bowl of the red and white salad dressings with some lettuce, when I began pondering the anagrams on my placemat.


Taking the Suck Out of Google Docs Documents

I thoroughly enjoy the flexibility of Google Docs. It's sufficiently fast, very accessible, and the simple but powerful sharing and chat features are, in my experience, second to none. However, despite Google's feature-rich and intuitive environment, I find in it one terrible flaw...


Weather is awesome!

Here's a taste of the brief storm that passed through Champaign at around 17:30:


Jonny and the Cell (pt. 2)

Specifically, what distance can do is act as a good rule of thumb. When searching for best practices regarding inter-access-point distance and hardware placement, I happened upon the following information, listed in the specifications for the equipment being used at a site I was looking into: