Ursa Labs and Apache Arrow in 2019
summary
A not-for-profit development group called URSA labs forms to steward Apache Arrow work with sponsors like 2 Sigma and Nvidia and a distributed team of five full-time engineers. The project defines a language-independent open standard for data frames, implements bindings for the arrow libraries (C++ core) and aims to memory map and interact with very large on disk datasets. It also plans to translate deep liar expressions into parallel runtime and use LLVM for function specialization and faster IO and an evolved Feather format.
AI-written summary · about
chapters (9)
- 0:00Start
- 0:27Ursa Labs Mission
- 0:53Ursa Labs Team
- 4:51Defragmenting Data Access
- 6:10Analytic database, deconstructed
- 6:43Arrow Use Cases
- 8:24Use cases for the R community
- 8:54Arrow's relationship with dplyr and friends
- 10:47What about Feather (format)?
related talks
33mArrow in Flight - New Developments in Data Connectivity
01h 02mApache Arrow: High-Performance Columnar Data Framework
45mApache Arrow: In Pursuit of Speed
34mApache Arrow as a full stack data engineering solution
39mTaming the language border (...)
45mAccelerating OpenSearch With Streaming: Apache Arrow, Flight, Data...
06mInterview
32mVectorized Query Processing for CPUs using Apache Arrow