BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Push versus Pull-Based Loop Fusion in Query Engines - Amir Shaikhh
 a\, Oxford
DTSTART:20190412T130000Z
DTEND:20190412T140000Z
UID:TALK118603@talks.cam.ac.uk
CONTACT:Victor Gomes
DESCRIPTION:Database query engines use push-based or pull-based approaches
  to avoid the materialization of data across query operators. In this talk
 \, we study these two types of query engines in depth and present the limi
 tations and advantages of each engine. Similarly\, the programming languag
 es community has developed loop fusion techniques to remove intermediate c
 ollections in the context of collection programming. We draw parallels bet
 ween databases (DB) and programming language (PL) research by demonstratin
 g the connection between pipelined query engines and loop fusion technique
 s. Based on this connection\, we propose a new type of pull-based engine\,
  inspired by a loop fusion technique\, which combines the benefits of both
  approaches. Then\, we experimentally evaluate the various engines\, in th
 e context of query compilation\, in a fair environment\, eliminating the b
 iasing impact of ancillary optimizations that have traditionally only been
  used with one of the approaches. We show that for realistic analytical wo
 rkloads\, there is no considerable advantage for either form of pipelined 
 query engine\, as opposed to what recent research suggests.
LOCATION:FW26
END:VEVENT
END:VCALENDAR