Performance problems doing analysis

[diane]

Steven noticed that doing an analysis seemed far slower than with original mussa, with his hox cluster using 3 20kbp regions. Several hours instead of 20 minutes.

(And we really do need a worker thread.)

Might be CPU/os dependent. though that seems pretty significantly different.

[diane]

Running with 1 sequence repeat masked

diane@brisby:~/proj/mussa_analyses/mck_1repeat/ChCoDoHuMoOpRa$ time ~/proj/mussa/build64/qui/mussagl --no-gui --run mck_w20_t14.mupa

real 7m44.508s user 7m43.182s sys 0m0.307s

---

[diane]

running with all repeat masked diane@brisby:~/proj/mussa_analyses/mck_repeatmasked/ChCoDoHuMoOpRa$ time ~/proj/mussa/build64/qui/mussagl --no-gui --run mck_w20_t14.mupa

real 3m36.728s user 3m35.672s sys 0m0.268s diane@brisby:~/proj/

[diane]

I loged out without copying no repeat masking was on the order of 13,500 minutes of real & user time.

[diane]

This ticket was mostly gathering information about what influences performance. Though it needs to be fixed, (but after a working release).