Reducing run times by breaking a problem into parts that can be completed by separate execution paths simultaneously.

learn more… | top users | synonyms

3
votes
2answers
69 views

How to reduce huge CXPACKET & LATCH_EX (ACCESS_METHODS_DATASET_PARENT) wait times?

Problem We have been experiencing high levels of user disruption due to SQL timeouts accross our systems since the beginning of the year. The SQL-Server instance in question has very high CPU usage ...
6
votes
1answer
80 views

Parallel option performing worse on cluster

I'm trying out oracles parallel option on a cluster and surprisingly I am getting worse results with the parallel option. I was expecting some improvement with the parallel option but certainly not ...
1
vote
1answer
50 views

Mathematical or Computing Alternatives for MySQL Combination Queries operating toward their Performance Limit

We regularly run a contest that is becoming quite popular. In the contest, we need to select the combinations of contestants from 8 groups whose cost is under a threshold, and who combined have the ...
3
votes
0answers
322 views

Are the following statements concerning sys.dm_exec_requests true?

Parallel plans mean that cpu_time can be greater than total_elapsed time. Therefore, per-query wait time is not possible to determine accurately from DMVs on servers with more than one CPU thread. ...
2
votes
0answers
37 views

Max Degree of Parallelism for SQL Server 2008

Microsoft have recommendations around setting MAXDOP on various versions of SQL Server, but this (old) article raises some issues and recommends setting MAXDOP = 1 generally for SQL Server 2005 and ...
2
votes
0answers
63 views

Performance differs on same query against two (almost) identical tables?

I'm running these two queries in SQL Server 2012: SELECT tpua.Field1, CAST(round(twg.Field2 * 2, 0) / 2 As Decimal(8,1)) As Field2, CAST((round((twg.Field3/10.0) * 2, 0) / 2) * 10 As int) As ...
2
votes
0answers
1k views

Oracle parallel query: finding out which Producer a Consumer is waiting for

One of our materialized view refresh's (part of a daily ETL routine) was hanging today; it usually completes within 5 minutes, but this time, I had to kill it after 23 hours. V$SESSION contained this ...
1
vote
0answers
79 views

Parallel insert in db2

I am trying to import from 8 files and insert into a single table in db2. I am running 8 .sql files that will do the task, but only one of them actually seem to be running and inserting into table. ...
1
vote
0answers
165 views

Why does removing a RID by creating a clustered index lose parallelism and slow down execution by 2X

If I remove a RID lookup by replacing 2 non-clustered indices (non unique) with 2 identical clustered indicies (non-unique) where there weren't any before, why does this query slow down by a factor ...
0
votes
0answers
5 views

DB with column store and parallel query execution?

I am looking for open-source alternatives to SQL Server Enterprise Edition for a mostly-read analytics workload. The main key features that I'm looking for are Column-oriented storage Parallel ...
0
votes
0answers
30 views

Turn on multi-threading in Oracle for ETL purposes

As you can see from this question, I was able to configure multi-threading in SQL Server and build a workflow in Informatica to multi-thread a job to move data faster into SQL Server. Multi-thread ...
0
votes
0answers
71 views

MySQL - Update vs. Replace Into — why is Update non-parallel, and Replace Into allows multi-threads?

In trying to optimize a query that does (Insert ... on duplicate key update) .... someone suggested I do "Replace Into" to optimize performance. I'm not 100% certain the difference other than ... ...
0
votes
0answers
277 views

Sql Server, a single threaded query spawns multiple threads (or internal parellelizaion during execution)

In Sql Server, when I run a procedure or a query that takes multiple minutes, sometimes spawns multiple threads. Even though there is no MAXDOP hint or necessity in the query, I see multiple instances ...
0
votes
0answers
363 views

MySQL Cluster lock wait timeout exceeded when trying to deal with huge amount of queries

I'm trying to deal with many connections and sometime I get following error: ERROR : Lock wait timeout exceeded; try restarting transaction The bigger number of queries the least number of ...