Yes. In this case, I would suggest configuring the column families that have very few rows to be in a separate locality group. You should be able to do this in the shell with the command: setgroups groupname=colf1,colf2,colf3 -t tablename
Here, groupname is an arbitrary name for the group; colf1, colf2, and colf3 are the column families with few rows; and tablename is the table name. After you create the locality group, you will need to compact the table for the change to take effect: compact -t tablename -w For each table, you can create multiple locality groups tailored to the access patterns of your data. There is some additional information about locality groups in the user manual: http://accumulo.apache.org/1.7/accumulo_user_manual.html#_locality_groups On Mon, Oct 5, 2015 at 8:25 AM, z11373 <[email protected]> wrote: > Hi Josh, > I see there are 4 tablet files for that table, and all of them are in range > from 730MB to 860MB in size. > For those column families that have problem, they are in 2 of those 4 > tablets. > They are only a few rows, but for those column families which have no > problem, they have millions of rows. > This makes me thinking if the slowness because it has to find those 'few' > rows among those 'gigantic' rows in that physical tablet file? > > Thanks, > Z > > > > -- > View this message in context: > http://apache-accumulo.1065345.n5.nabble.com/scan-command-hung-tp15286p15320.html > Sent from the Developers mailing list archive at Nabble.com. >
