Tutorial Will Cover Critical Decisions and Experience Using MemSQL to Support Millions of Analytical Queries in Real Time
SAN JOSE, CA – March 13, 2017 – MemSQL, provider of the fastest database platform for real-time analytics, today announced an end-user presentation, “Real-time Analytics at Uber Scale”, at Strata+Hadoop World in San Jose. James Burkhart, technical lead on real-time data infrastructure at Uber, will present a tutorial session covering the critical decisions and lessons learned building an architecture for its Apollo Project that supports millions of analytical queries daily in real time.
Uber to Present on Real-Time Analytics at Strata+Hadoop World 2017 San Jose – Click to Tweet
Strata+Hadoop Session Details
Real-time Analytics at Uber Scale
James Burkhart, technical lead on real-time data infrastructure, Uber
11:00am–11:40am Wednesday, March 15, 2017
Location: LL20 B
The Apollo project at Uber was built to support millions of analytical queries daily across real-time data. During his presentation, James will cover the architectural decisions and lessons learned building an exactly-once ingest pipeline storing raw events across in-memory row-storage and on-disk columnar-storage, and a custom metalanguage and query layer leveraging partial OLAP result set caching and query canonicalization. Putting all the pieces together provides thousands of Uber employees with subsecond p95 latency analytical queries spanning hundreds of millions of recent events.
MemSQL and Uber
Uber deployed MemSQL to complement existing systems infrastructure and deliver real-time analytics at scale. In a recent blog post, Ringing in 2017, Uber showcased groundbreaking results for its busiest time of the year, including a series of stunning geographic visualizations produced by the Apollo project. During the New Year’s eve period, Uber provided around 15 million safe rides with an average ETA for a ride of just four minutes. The graphic for the San Francisco Bay Area shows how those rides transpired across that region.
Image source: Uber
MemSQL provides the real-time geospatial analytics infrastructure for Apollo. By taking advantage of the MemSQL distributed architecture and the platform’s real-time data warehousing capabilities, Uber was able to create a comprehensive picture of its business that is instantly up to date. Uber sends real-time inputs to MemSQL, which are automatically captured and deduplicated as needed with the transactional properties of the system, including unique keys. Combined with native geospatial functions, Uber can architect dashboards for the company down to individual city specific views.
Connect with MemSQL at Strata + Hadoop World 2017 San Jose
March 14-16, stop by MemSQL Booth #1019 for demos, games, and giveaways: memsql.com/strata
MemSQL delivers the leading database platform for real-time analytics. Global enterprises use MemSQL to achieve peak performance and optimize data efficiency. With the combined power of database, data warehouse, and streaming workloads in one system, MemSQL helps companies anticipate problems before they occur, turn insights into actions, and stay relevant in a rapidly changing world. Visit memsql.com or follow us @memsql.
Flak42 for MemSQL