hadoop - Can someone explain unique coprocessor execution behavior we are noticing? -


our environment :

  • we have 44 region servers,
  • the table has 572 regions,
  • each region server has 13 regions
  • each region has anywhere between 40k 110k records
  • each region server 32 core/190gb ram

my issue when run coprocessor runs on these regions radically different execution time on regions of single region server.

for e.g. 1 region took 3.9 seconds whereas took 8.9 seconds on same region server same no. of rows.

is normal behavior? region coprocessor execute under single jvm or each region has own handler in own jvm?

what significance of hbase.regionserver.handler.count ? set 100. mean 100 threads in single jvm or multiple jvms?

got answers of questions:

is normal behavior? region coprocessor execute under single jvm or each region has own handler in own jvm?

all regions handled single jvm process.

what significance of hbase.regionserver.handler.count ? set 100. mean 100 threads in single jvm or multiple jvms?

this max no. of threads spun in hbase jvm process handling requests particluar regions.+


Comments

Popular posts from this blog

php - Submit Form Data without Reloading page -

linux - Rails running on virtual machine in Windows -

php - $params->set Array between square bracket -