CAP and you

Consistency…!Availability…!Messages…!

& You…

CAP!theorem

Consistent “…as if every request was executing on a single node, responding to operations one at a time.”

ACID Compliant

Available “…every request received by a non-failing node

in the system must result in a response.”

Partition Tolerant “…the network will be allowed to lose arbitrarily many messages sent from one node to another”

Pick Any Two!?!! It’s complicated…

Consistent & Partition Tolerant “…it is easy to achieve atomic data and partition

tolerance. The trivial system that ignores all requests meets these requirements.”

Consistent & Available “If there are no partitions, it is clearly possible to

provide atomic, available data.”

Available & Partition Tolerant “If there are no consistency requirements, the

service can trivially return v0, the initial value, in response to every request.”

€

2 ⋅ tmessage + tlocalt-Connected Consistent

What has been neglected?  Latency  Failure  Computation  Durability  More™

Latency Petite Partitions

the little deaths

When the available consistency exceeds the required availability a choice must be made…

Raising the question, how are business decisions made?

What protocols are used to pass messages in your organization?

Has your organization

chosen consistent, available

or partition tolerant?

In any human interaction, the required amount of communication is inversely proportional to the level of trust…

@bhorowitz

Andrew Clay Shafer @littleidea

work on stuff that matters