DNA and RNA of Ruptured Computation
First, I introduce the notion of a ruptured computation that is a relaunched, restarted or resumed process from a checkpoint after a rupture in its fabric, for example, crash, hang, spike and leak forcing termination. DNA can be an abbreviation of a Dump of Nasty Application or Double-ruptures of Nasty Application but I got an enlightenment spike in my head recently after recognizing that C, H, S and L are 4 DNA bases: Crash, Hang, Spike and Leak. Do you see a parallel with the real DNA and its 4 bases: A, T, G, C? Now a sample of a ruptured computation DNA strand (imagine, we launch an application and it crashes, …, crashes, …, crashes, …, spikes, …, crashes, …, hangs, …) can be represented as: …CCCSCH… . What about its second strand? I’m in search of it now but in the mean time I introduce the second notion of RNA (Really Nasty Application or Ruptures of Nasty Application) that is just one strand of …CCCSCH… as needed (real RNA is mostly single-stranded).
- Dmitry Vostokov @ DumpAnalysis.org -
October 24th, 2009 at 1:30 am
I was thinking about something kinda similar to this today, and this post brings some toughts to me. By adding an abstract/semantical layer and simplyfying the inherent complexity of debuggers following the CHSL criteria that you wisely mentioned and reagrouping the most common anomalies in software functioning (crash, hang, spike and leak in this case) by some basic heuristic, I think there’s a possibility to create a graphical debugging and analysis framework; just think about finding patterns in a software instructions flow, like EEELC (E stands for Expected result, L for Leak and C for Crash) by looking at a graph or any other data visualization method, i.e, among other possibilities.
Just a thought, I hope I’d make my point clear.
October 27th, 2009 at 10:36 am
Sounds interesting! Will think about it too
Thank you,
Dmitry
October 31st, 2009 at 2:03 pm
You’re welcome.
: )