LIKE causing delay in query execution

Question

Hi,
I have this query that takes about 20 mins to execute. Is there a way to tune this, especially for the LIKE where clauses?
SELECT product_id, territory_code, global_target_id,
DECODE (service_type, 'SERVICE', 'Y', 'N') service_type,
equipment_deployment, created_date, updated_date,
fiscal_period_id, customer_id,
SUM (bookings_amount) bookings_amount, ship_to_country,
alliance_id
FROM sp_refresh_incre_vw2@cdw
WHERE fiscal_period_id = in_chr_fiscal_period_id
AND ( ec_name LIKE in_chr_search_string --- culprit
AND NOT st_name LIKE in_chr_search_string --- culprit
AND NOT bt_name LIKE in_chr_search_string --- culprit
GROUP BY fiscal_period_id,
product_id,
territory_code,
ship_to_country,
customer_id,
global_target_id,
service_type,
equipment_deployment
HAVING SUM (bookings_amount) <> 0
UNION
     SELECT product_id, territory_code, global_target_id,
DECODE (service_type, 'SERVICE', 'Y', 'N') service_type,
equipment_deployment, created_date, updated_date,
fiscal_period_id, customer_id,
SUM (bookings_amount) bookings_amount, ship_to_country,
alliance_id
FROM sp_refresh_incre_vw2@cdw
WHERE fiscal_period_id = in_chr_fiscal_period_id
AND bt_name LIKE in_chr_search_string --------- culprit
GROUP BY fiscal_period_id,
product_id,
territory_code,
ship_to_country,
customer_id,
global_target_id,
service_type,
equipment_deployment
HAVING SUM (bookings_amount) <> 0
UNION
SELECT product_id, territory_code, global_target_id,
DECODE (service_type, 'SERVICE', 'Y', 'N') service_type,
equipment_deployment, created_date, updated_date,
fiscal_period_id, customer_id,
SUM (bookings_amount) bookings_amount, ship_to_country,
alliance_id
FROM sp_refresh_incre_vw2@cdw
WHERE fiscal_period_id = in_chr_fiscal_period_id
AND ( st_name LIKE in_chr_search_string -------- culprit
AND NOT bt_name LIKE in_chr_search_string -------- culprit
GROUP BY fiscal_period_id,
product_id,
territory_code,
ship_to_country,
customer_id,
global_target_id,
service_type,
equipment_deployment
HAVING SUM (bookings_amount) <> 0;
After some investigation, I have noted that it is the LIKE clause that is causing the delay. The three queries are exactly similar EXCEPT for the LIKE clause, which I have commented on with "--- culprit"
Any alternative ideas to achieve a similar result by avoiding the LIKE clause will be very much appreciated.
Thanks and regards,
Ambili

Answer 1

Hi,
.I Accelerator BI accelerator is a separate hardware which when attached to BW server creates an index on data of infocube and stores in compressed format.Different technologies are used for compression and horizontal pratitioning of data. BI Accelerator can benefit businesses that have high volumes of data.
This link explains you better on BI Accelerator
What is BI Accelerator
Comming to the query performance, Using BIAccelerator, query performance is highly improved.
There is o ideal time for the query execution time.
The query execution time depends on various factors, to list some of them,
1)Complexity of the query(more complex, more execution time)
2)Object on which the query is built (If it is built on CUbe, Performance will be more, on Dso performance is poor)
3)Building indexes on cube increases performance....
4)Data volume in the report.
But yes, We can look at the stats of query execution using ST03 and ST03N .
Hope the above answer helps you.
Please let me know if you need any more clarifications