Hi, You can try to 1. add more nodes to reduce data set on each node. 2. use query parallelizm [1]. 3. add a indexed field to be used in WHERE clause instead of function.
[1] https://apacheignite.readme.io/docs/sql-performance-and-debugging#query-parallelism On Tue, Aug 29, 2017 at 12:09 PM, mhetea <mihaela.he...@gmail.com> wrote: > We have a cache with ~3million entries > > We execute the following query: > select count(_key) from IgniteProduct as product where POSITION > ('Z',manufacturerCode)>0 > > The query takes about 7s though the index is hit. > > Query execution is too long [time=7103 ms, sql='SELECT > COUNT(PRODUCT__Z0._KEY) __C0_0 > FROM "productCache".IGNITEPRODUCT PRODUCT__Z0 > WHERE POSITION('Z', PRODUCT__Z0.MANUFACTURERCODE) > 0', plan= > SELECT > COUNT(PRODUCT__Z0._KEY) AS __C0_0 > FROM "productCache".IGNITEPRODUCT PRODUCT__Z0 > /* "productCache".IGNITEPRODUCT_MANUFACTURERCODE_IDX */ > WHERE POSITION('Z', PRODUCT__Z0.MANUFACTURERCODE) > 0 > , parameters=[]] > , parameters=[]] > > Can somebody help/offer an alternative? > > > > -- > View this message in context: http://apache-ignite-users. > 70518.x6.nabble.com/SQL-query-is-slow-tp16475.html > Sent from the Apache Ignite Users mailing list archive at Nabble.com. > -- Best regards, Andrey V. Mashenkov
