I deploy TPC-H test database (scale factor 50) on my test virtual machine (SLES 11SP3, 16VCPU, 48GB RAM)
on HANA SPS09 rev. 96 database.
Linear fullscan performance is amazing, but for query's with join's operation got some not optimal performance
For example (little changed 3.sql script from TPC-H) :
select TOP 100
sum(l_extendedprice * (1 - l_discount)) as revenue,
c_custkey = o_custkey
and l_orderkey = o_orderkey
and c_mktsegment = 'AUTOMOBILE'
Tables rows count :
LINEITEM - 300 M
ORDERS - 75 M
CUSTOMER 7.5 M
Sustained average execution time of this query is 35 sec.
I try to create indexes on join columns, create statistics on all tables, use some JOIN-hint's, but with no luck
I collect and attach EXPLAIN_PLAN, PlanViz graph, and output of "HANA_SQL_StatementHash_KeyFigures.txt" script.
Will be very appricated if someone can give any ideas for improvment of this query