BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//sf26-agenda//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
BEGIN:VEVENT
UID:SESS-31@sf26-agenda
DTSTAMP:20261020T151709Z
DTSTART:20261104T210000Z
DTEND:20261104T214500Z
SUMMARY:Sizing\, Benchmarking and Performance Tuning Apache Flink Clusters
LOCATION:Room 2008
DESCRIPTION:Robert Metzger\n\nA common question when adopting Apache Flink 
 is about sizing the workload: How many CPUs\, how much memory will Flink r
 equire for a particular use-case? What throughput and latency can you expe
 ct given your hardware? We’ll kick off this talk discussing why these qu
 estions are extremely difficult to answer for a generic stream processing 
 framework like Flink. But we won’t stop there. The best approach to answ
 er sizing questions is to benchmark your Flink workload. We will present h
 ow we’ve set up a Flink SQL-based benchmarking environment and some benc
 hmarking results for attendees to correlate our results with their workloa
 ds to approximate their resource requirements. Naturally when benchmarking
 \, the topic of performance tuning comes up: Are you optimally using the a
 llocated resources? How to identify performance bottlenecks? What are the 
 most common performance issues\, and how to resolve them? In our case\, a 
 few configuration drastically improved the performance. How many CPU cores
  are needed for that in Flink? Attend the talk to find out\, less than you
  would expect. This talk is for both Flink beginners wanting to get an ide
 a about Flink’s performance and operational behavior\, as well as for ad
 vanced users looking for best practices to improve performance and efficie
 ncy.
URL:https://sf26.rmoff.net/s/sizing-benchmarking-and-performance-tuning-apa
 che-flink-31
END:VEVENT
END:VCALENDAR