~/talk

dShark: A General, Easy to Program and Scalable Framework for Analyzing In-network

Talkat NSDI 2019 · 2019-02-26..02-28· [networking]
name:Da Yu
links:none on file
name:Yibo Zhu
links:none on file
links:none on file
links:none on file
links:none on file
name:Karl Deng
links:none on file
links:none on file
slides:none on file

dShark provides a MapReduce-like pipeline that parses mirrored packets into compact summaries using a JSON declarative spec and runs per-group analysis with C++ per-group UDFs, correlating instances despite middlebox-added headers. It tolerates capture noise by using TCP-seq evidence, implements 18 ready apps, and processes >10 Mpps on a single commodity server with near-linear scale to ~26.4 Mpps (16 passers, 24 groupers). Drop-localization reduces false positives from ~5,600 to 7.

AI-written summary · about

  1. 0:00Intro
  2. 0:32Network reliability is critical
  3. 0:55Existing tools are the first attempts
  4. 2:17In-network packet capture is the last resort
  5. 3:13In-network analysis: challenges
  6. 3:49Example: Route error checker
  7. 5:32Packet headers are modified by middleboxes
  8. 6:55dShark: three goals
  9. 7:30How operators manually process traces
  10. 9:53dShark's programming model
  11. 11:16Declarative spec. in parsers and groupers
  12. 12:20Same protocol headers bring ambiguity
  13. 13:11Imperative diagnosis logic in the query processors
  14. 14:02dShark overview
  15. 14:35Another example: load balancer profiler
  16. 16:09Tolerate capture noise
  17. 17:21Performance of Shark
  18. 20:24Conclusion
~/search
type to search 6,249 conferences, 138,987 speakers and 185,172 talks
~/help
cmdctrl k
search every conference, speaker and series
/
focus the filter search
h
home
c
conferences
s
speakers
e
series
t
light / dark — the status bar button does the same
?
this help
esc
close this, then clear every filter — or just the search, while you are typing in it

Ignored while you are typing. Pages without a filter rail have no search to focus.