Analyze and tune map reduce job is another thing you might try. On Thu, Mar 31, 2016 at 2:49 PM, 钱巧能 <[email protected]> wrote:
> The time cost of cubing job depends on many things, like dataset size, > hadoop capacity and your cube design etc. And what Kylin can help is cube > design, you can refer to > https://kylin.apache.org/docs15/howto/howto_optimize_cubes.html > & https://kylin.apache.org/blog/2016/02/18/new-aggregation-group/ to > optimze your cube design. > > > > At 2016-03-31 11:53:21, "Bin.Li" <[email protected]> wrote: > > hello, > > It takes too much time to build a cube(hours). > > Less than 10 dimisions & less than 5 measures. > > bin.li > > > > >
