Our number 1 value at Snowflake is to “Put Customers First”. Whether it’s through major product releases, feature enhancements, or documentation, we wholeheartedly embrace this value at Snowflake. Throughout 2018, we’ve been working on a special holiday gift to show our gratitude for all of our customers. It’s a gift every customer will be able to appreciate. It’s the gift of increased performance.
Faster access to data
Over the past year, the Snowflake engineering team worked diligently on improving the performance of Snowflake — all at zero operational or disruption costs to customers. Each week the team delivered significant feature enhancements which favorably impacted query workloads, including:
- Query processing improvements
Snowflake continuously makes improvements to query performance. Some of our big improvements this year include better micro-partition pruning to improve join performance, query optimizations on window functions, and improvements to table sampling.
- Upgrades to the underlying hardware
By updating the type of servers that Snowflake runs on, we’ve ensured a faster CPU, a notable increase in network bandwidth, and an increase in overall storage performance. - Materialized views
Materialized views improve query performance of workloads composed of common, repeated query patterns, and they’re all automatically and transparently maintained by Snowflake. - Auto-clustering
Auto-clustering is Snowflake’s re-clustering service that provides automated and optimized self-organization of data storage, removing the burden of manually re-clustering data.
Measuring success against the highest standards
We measured and verified these performance improvements by utilizing an industry-standard benchmark workload: A 10TB version of a query set from the TPC-DS benchmark. The TPC Benchmark DS (TPC-DS) is a decision support benchmark that models several generally applicable aspects of a decision support system, including queries and data maintenance.
The gift of speed and efficiency
We only succeed when our customers succeed, so this holiday season we hope you enjoy our gift of increased performance. Our goal is to continue helping data analysts and business intelligence teams do their jobs more effectively and efficiently.
But, don’t take our word for it. Customers like Jay White, Director of Software Engineering at Blackboard, describes the experience of having faster performance in his own words. “Everything we did with Snowflake left our jaw on the table,“ White said. „We’ve never experienced anything like it before, it just runs incredibly fast.” Click here to learn more about materialized views, auto-clustering, and more.
Happy Holidata from all of us at Snowflake!