Dramatis Personae
Serge Danzanvilliers (beard)
@alolcool
Yann Schwartz (no beard)
@abolibibelot
Slide 3
Slide 3 text
2009
We got some data
Slide 4
Slide 4 text
No content
Slide 5
Slide 5 text
2010
We got mucho data
Slide 6
Slide 6 text
No content
Slide 7
Slide 7 text
A distributed system is one in which the failure
of a computer you didn't even know existed can
render your own computer unusable.
Leslie Lamport
Slide 8
Slide 8 text
2012
Wait what’s this data anyway?
Slide 9
Slide 9 text
No content
Slide 10
Slide 10 text
No content
Slide 11
Slide 11 text
No content
Slide 12
Slide 12 text
2013
Invented Elsewhere
Slide 13
Slide 13 text
Kafka!
●
Clustered
●
Failover
●
Replication
●
Non texty
●
Pull model
●
Streaming
Slide 14
Slide 14 text
No content
Slide 15
Slide 15 text
No content
Slide 16
Slide 16 text
(Interlude)
The Case of the Missing Minutes
Slide 17
Slide 17 text
No content
Slide 18
Slide 18 text
No content
Slide 19
Slide 19 text
Reality has a known theoretical bias
Slide 20
Slide 20 text
No content
Slide 21
Slide 21 text
Buffers, buffers everywhere
●
Caches, buffers, batches, queues : everything comes in chunks
●
Unbounded queues : never-ending tears
●
Operation triage: lose some data or lose your system
Slide 22
Slide 22 text
TL;DR
●
Ignorance is not bliss
●
System entropy always increases
●
Worse is better until it’s not