Item: Treasure Data
Rating: 9
Author: Igor Pernek

Overall Satisfaction with Treasure Data

Use Cases and Deployment Scope

It is used as a data warehouse. Raw data containing operative business information is continuously ingested to Treasure Data. Different batch jobs are continuously running to aggregate data based on a different number of dimension to achieve relevant granularities for specific use cases. The data is used across the whole organisation for: reporting to customers, troubleshooting by tech support, analysis and feature design by R&D, etc.

Pros and Cons

Pros

Very reliable data ingestion to their warehouse
Fast querying capabilities using Presto and Hive
Good API for programatic access to the data

Cons

Adding support for not only batch processing but also streaming data
Better grouping and organization of saved queries and jobs

Return on Investment

It was positive to not need to implement a data warehousing solution ourselves
It allowed us to efficiently store important business (e.g.billing) information in real time

Innovative Uses

For implementing revenue boasting algorithms based on historic data
For implementing cost saving algorithms using historic data

Other Software Used

Docker, Aerospike Database, MemSQL

Likelihood to Recommend

It is very well suited for use cases where the main requirement is to process store data in a scalable warehouse as a service and do batch processing of the data. It is less appropriate if the main business requirement is to be able to process streaming data in real time.

Comments

Please log in to join the conversation

Reliable data warehouse as a service