Hi David, Since join is implemented through hash-join and table stats are currently unavailable yet to help better decide the join order (and which tables should be small enough to go into memory), users may sometimes have to be careful about the table order in the query. In your case, you can try moving the lineitem table (which is the largest here) to the first place in the query. Think that should work.
Thanks, Maryann On Mon, Feb 24, 2014 at 6:08 PM, David Wang <[email protected]> wrote: > Hi Maryann, > > The sizes of each table in my query are as follows: > customer - 24 MB > orders - 164 MB > lineitem - 725 MB > supplier - 1.4 MB > nation - 2.2 KB > region - 400 B > > The heap size of my region servers is 4 GB. > > Thank you, > David > > > On Mon, Feb 24, 2014 at 5:58 PM, Maryann Xue <[email protected]>wrote: > >> Hi David, >> >> What are sizes of each table in your query? And what is the heap size of >> your region server? >> >> >> Thanks, >> Maryann >> >> >> On Mon, Feb 24, 2014 at 5:50 PM, David Wang <[email protected]>wrote: >> >>> Hi, >>> >>> I had a question about running queries containing inner joins on >>> phoenix, but wasn't sure where to post. >>> I downloaded the phoenix-master (3.0.0), and inserted 10 MB of data >>> through psql.sh. I found I was able to run a query, >>> which contains an inner join in sqlline. >>> >>> The query I tried: >>> >>> select n_name, sum(l_extendedprice * (1 - l_discount)) as revenue from >>> customer inner join orders on c_custkey = o_custkey inner join lineitem on >>> l_orderkey = o_orderkey inner join supplier on l_suppkey = s_suppkey and >>> c_nationkey = s_nationkey inner join nation on s_nationkey = n_nationkey >>> inner join region on n_regionkey = r_regionkey where r_name = 'AMERICA' and >>> o_orderdate >= '1993-01-01' and o_orderdate < '1994-01-01' group by n_name >>> order by revenue desc; >>> >>> But when I increased the data size from 10 MB to 1 GB, and try to run >>> the same query, I get the following error: >>> Error: Encountered exception in hash plan execution. (state=,code=0) >>> >>> I have included the path to snappy-java-1.0.3.2.jar in sqlline.sh like >>> the following: >>> java -cp >>> ".:$phoenix_client_jar:/usr/local/hbase-0.94.8/lib/snappy-java-1.0.3.2.jar" >>> -Dlog4j.configuration=file:$current_dir/log4j.properties >>> sqlline.SqlLine >>> -d com.salesforce.phoenix.jdbc.PhoenixDriver -u jdbc:phoenix:$1 -n none >>> -p >>> none --color=true --fastConnect=false --verbose=true >>> --isolation=TRANSACTION_READ_COMMITTED $sqlfile >>> >>> Do you have any idea where the problem might be? >>> >>> Thank you so much, >>> >>> David >>> >> >> >> >> -- >> Thanks, >> Maryann >> > > -- Thanks, Maryann
