The job is distributed, but each task is single threaded. That's the MapReduce model :)
J-D On Mon, Sep 26, 2011 at 4:45 PM, Rita <[email protected]> wrote: > thanks. I noticed the MR job is single threaded. Is table export a serial > based job or can it be done in parallel? > > > > On Mon, Sep 26, 2011 at 12:41 PM, Jean-Daniel Cryans > <[email protected]>wrote: > >> If by "export" you refer to the Export MR job that's shipped with >> HBase, then it's not going to work at all since the table is disabled. >> >> If you refer to a distcp, which would make more sense since by being >> disabled the table won't get new mutations, then it won't be any >> faster... just safer. >> >> J-D >> >> On Mon, Sep 26, 2011 at 2:03 AM, Rita <[email protected]> wrote: >> > Is there a performance boost to export a table while its disabled? >> > >> > -- >> > --- Get your facts first, then you can distort them as you please.-- >> > >> > > > > -- > --- Get your facts first, then you can distort them as you please.-- >
