YeSQL You extend SQL with Rich and Highly Performant User-Defined Functions in Relational Databases

Posted on July 26, 2023   1 minute read ∼ Filed in  : 

Introduction

Gap

Python UDF is not using JIT compiler. And the funciton

Goal

This paper try to improve the performance by

  1. reducing data conversions, copies, and context switches between different execution environments, when UDF input and output is translated from and to SQL.
  2. overhead of running Cpython’ interpreter.

Techniques:

  1. seamless data exchange between the UDF and the DBMS.
  2. JIT-compiled UDFs.
  3. UDF parallelization
  4. activation patterns, and number of dichotomies.stateful UDFs, and UDF fusion

CFFI enable python to call C code,

s

easy implementation of complex algorithms and several performance enhancements, including tracing JIT compilation of Python UDFs, parallelism and fusion of UDFs, stateful UDFs seamless integration with a database engine




END OF POST



Tags Cloud

Categories Cloud

   A paper note    CMU database    coding    devices    distributed database    machine learning basic    operation system    practise    programming language   


It's the niceties that make the difference fate gives us the hand, and we play the cards.