refactored SharedHashMapTest for easier understanding

peter-lawrey / HugeCollections-OLD

Huge Collections for Java using efficient off heap storage

273 stars 51 forks source link

refactored SharedHashMapTest for easier understanding #8

Closed JozsefBartokHFT closed 10 years ago

JozsefBartokHFT commented 10 years ago

Pls. take a look at this. I've done this changes because I found the test very hard to understand in its original form. Since I've removed some lines I didn't see the point of I might have broken behavior. Pls. take a look.

Cotton-Ben commented 10 years ago

For SHM 200M tests , when you brIdge from on-heap view to a 'real' UFS (eg /ext4) operand custody view, is that why the test's final /proc/$PID/status VMsz: (Et. Al. VMxxx:) metrics are so much smaller than the CHM 200M test's /proc/$PID/status. VMsz; metric result?

Does this somehow imply that the SHM@/ext4 data structure is both OffHeap and OffPID?

If not, what does account for the big diff between SHM 200m@/ext4. Vs. CHM 200m@onHeap?

Great work u guys!