Great Analytic Data-Warehouse
June 13, 2019

Great Analytic Data-Warehouse

Anonymous | TrustRadius Reviewer
Score 10 out of 10
Vetted Review
Verified User
Review Source

Overall Satisfaction with Amazon Redshift

Amazon Redshift is being used as our primary analytic data-warehouse. This allows our data and analytic team to build report and query data without going directly to our production database. It is a central data repository from external data sources as well, data we import from 3rd parties and segment.
  • Complex queries
  • Aggregation
  • Fully managed service
  • Works very well with most BI/reporting solutions
  • Stored procedures
  • Job scheduling
  • A easier way (perhaps a GUI) to manage users permission
  • Easy to set up and very cost effective to run large complex queries.
  • Separation of roles between our production and analytic environment.
  • A lot of cloud service is RedShift ready (i.e., Segment, FiveTran, Matillion) which decreases implementation efforts
Most of our stack is on AWS, so while Snowflake and BigQuery was a viable option from a performance perspective, it was easier to integrate with RedShift. We considered hosting SQL Server on AWS or using Amazon RDS (Postgres or MySQL), however, the self-service aspect of RedShift and its superior parallel query execution made us choose RedShift.
RedShift is great for analytics, dealing with a large volume of data and running complex queries. The only reason why it is a 9 and not a 10 is the lack of procedural capabilities out of the box, which requires another software (Matillion, FiveTran) or AWS service (Lambda, Data Pipeline, Kinesis) to fully unleash the power of RedShift.
Amazon Redshift is great for analytics, reporting, and complex queries for statistical modeling and machine learning. Its ability to run parallel queries in a simple SQL environment makes the transition from traditional DB very easy. Very good for loading/reading/writing large datasets. I would not recommend RedShift for an environment that requires single row reads/updates, which it is not optimized for.