Analysis between BiqQuery and Snowflake for EDP

Analysis is done on the basis of different criteria.

• Pricing
• Ease of Use
• Portability
• Performance
• Scalability

1. Pricing: BigQuery costs for every query. Whereas, in Snowflake you pay for how much compute     takes place and you can run any number of queries without any charges.

2. Portability: BigQuery is a native Google Cloud Platform (GCP) service so it only runs on the Google Cloud. Snowflake is available for AWS and Azure and they just announced they are working on a port for the Google Cloud.

3. Performance: Speed- Snowflake is faster than BigQuery

• In a head-to-head test, Snowflake edged out BigQuery in terms of raw speed, with queries taking, on average, 10.74 seconds (geometric mean). Meanwhile, BigQuery clocked in at 14.32 seconds per query, on average.
• According to independent third-party benchmarks, Snowflake performance is noticeably better than BigQuery performance.

4. Ease of Use: Both Snowflake and BigQuery score highly on the usability scale, although Snowflake may be slightly easier to use. In particular, BigQuery’s serverless nature makes it easy to get up and running quickly.

5. Scalability: Snowflake and BigQuery both have advanced scalability features. However, BigQuery comes out slightly ahead by handling everything under the hood, removing the need for users to perform any manual scaling or performance tuning.

Conclusion:

• According to its creators, Snowflake is “faster, easier to use, and far more flexible than traditional data warehouse offerings.” Rather than building on existing solutions such as Hadoop, Snowflake uses a new SQL database engine with a cloud-optimized architecture.

• Snowflake is also noteworthy because it enables total separation between a data warehouse’s compute and storage requirements. This allows you to scale both requirements up and down independently, giving you greater flexibility while cutting costs.
• In a series of 2019 benchmark tests, Technology blog GigaOm found that Snowflake consistently outperformed BigQuery on a number of metrics.
• Snowflake and BigQuery are both feature-rich data warehouse solutions that have helped companies of all sizes and industries improve their BI and analytics workflows.
• Although BigQuery is cheaper than Snowflake in terms of storage, BigQuery’s unique query-based pricing model is very different from Snowflake’s time-based pricing model with multiple compute tiers. In addition, Snowflake generally beats BigQuery in terms of performance, according to the latest set of benchmarks.

• So, based on our EDP Pipelines and Usage Snowflake will be the better choice as snowflake charges only for the compute and not for each query, whereas Big Query charges per query.