Monday, August 19, 2019

AWS Redshift Performance versus SQL Multidimensional and Tabular cubes

This is an update to my recent blog AWS Redshift versus SQL Server cube. This performance benchmark includes SQL Server's Tabular cube technology, which wasn't possible as the in-memory database required more memory than available in the test server.
This updated benchmark includes
  • AWS Redshift
  • SQL Columnstore
  • SQL Multidimensional Cube
  • SQL Tabular Cube

The cubes were the slam dunk winners, which shouldn't surprise any Data Platform Architects. For many Data Platform Architects, this will validate their use of columnar databases for atomic data combined with an OLAP cube for fast and scalable BI queries.

The races within the race were interesting. With Multidmensional beating Tabular by a factor of two, and SQL Columnstore beating AWS Redshift by a factor of two also.

For the full report goto AWS Redshift versus OLAP Cubes Performance Benchmark AWS Redshift versus OLAP cubes: Performance Benchmark.

3 comments:

joeharris76 said...

Thanks for publishing the table definitions. I note that you did not define a distribution key on the fact tables. This will require Redshift to move quite a bit of data around to provide your answers.

Regarding the query timings, I note that many of your queries (particularly Q5) make use of correlated subqueries. This is an anti-pattern for Redshift. Correlated subqueries become very expensive in an MPP system like Redshift. The correlated subquery can almost always be rewritten to use an outer join.

I'm confident that fixing these 2 issues would _dramatically_ improve the Redshift timings. In my personal experience Redshift is 10-100x faster than an equivalent SQL Server setup.

Unknown said...

Hi, that was an interesting article! A couple points -
1. Distribution keys based on query usage is key for Redshift performance. I agree with the person above that if we choose appropriate distribution keys, it will make a significant difference.
2. Cubes in SQL Server have already pre-computed a lot of the query and stored the results in an optimized manner. It would be interesting to compare Microstrategy cubes to SQL Server cubes.

Bhanu Sree said...

You have done a great job, There are may person searching about that topic. now they will easily find your post
RedShift Training
AWS RedShift Training
Amazon RedShift Online Training