[reading review] Velox: Meta's Unified Execution Engine

In this paper, the authors design Velox, an extensible C++ library to accelerate database query execution. Velox does not provide SQL parser, query optimizer, and other traditional DBMS components. It takes physical plans as input and produces output for queries. Velox provides consistent, reusable and high-performance data processing components to data users. It does not own data itself, but can deal with different existing data format as input.

  • Strengths: Velox is highly extensible, reusable and high-performance, which makes it easy to maintain and do further development.

  • Future works: Velox can do further development to integrate with ML platforms and can do more optimizations during adaptive query execution.




Enjoy Reading This Article?

Here are some more articles you might like to read next:

  • [reading review] MonetDB/X100: Hyper-Pipelining Query Execution
  • [reading review] Lakehouse: A New Generation of Open Platforms that unify Data Warehouse and Advanced Analytics
  • [reading review] The FastLanes Compression Layout: Decoding >100 Billion Integers per Second with Scalar Code
  • [reading review] Exploiting Cloud Object Storage for High-Performance Analytics
  • [reading review] OceanBase: A 707 Million tpmC Distributed Relational Database System