Scribd
Upload
50 views

CAP Theorem

This document discusses the CAP theorem, which states that it is impossible for a distributed computer system to simultaneously provide consistency, availability, and partition tolerance. It defines these three terms, explaining that only two of the three guarantees can be achieved at the same time. For example, a system can prioritize either consistency and availability (when there is no network partition), or availability and partition tolerance (when consistency is relaxed during network issues). The document also discusses the BASE approach of having systems that are Basically Available, Soft-state, and Eventually consistent.

Uploaded by

Alex T
Copyright
© Attribution Non-Commercial (BY-NC)
Available Formats
Download as PDF, TXT or read online on Scribd
50 views

CAP Theorem

This document discusses the CAP theorem, which states that it is impossible for a distributed computer system to simultaneously provide consistency, availability, and partition tolerance. It defines these three terms, explaining that only two of the three guarantees can be achieved at the same time. For example, a system can prioritize either consistency and availability (when there is no network partition), or availability and partition tolerance (when consistency is relaxed during network issues). The document also discusses the BASE approach of having systems that are Basically Available, Soft-state, and Eventually consistent.

Uploaded by

Alex T
Copyright
© Attribution Non-Commercial (BY-NC)
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 28

CAP Theorem

Grounds of Despair

Sunday, 22 September 13

YEAR 2000

Sunday, 22 September 13

Professor Eric A. Brewer

Sunday, 22 September 13

Professor Eric A. Brewer

First who ever said..

Sunday, 22 September 13

IMPOSSIBLE TO BUILD STABLE SYSTEM

Sunday, 22 September 13

Sunday, 22 September 13

...it is impossible for a distributed computer system to simultaneously provide all three of the following guarantees:..

Sunday, 22 September 13

Consistency all nodes see the same data at the same time

Sunday, 22 September 13

Availability a guarantee that every request receives a response about whether it was successful or failed

Sunday, 22 September 13

Partition tolerance the system continues to operate despite arbitrary message loss or failure of part of the system

Sunday, 22 September 13

Consistency Partition tolerance Availability

Sunday, 22 September 13

CAP Theorem

Sunday, 22 September 13

CAP Theorem

for dummies

Sunday, 22 September 13

Consistency

A service that is consistent operates fully or not at all

Sunday, 22 September 13

Consistency

System is not atomic

Sunday, 22 September 13

Availability

A service that is fully available

Sunday, 22 September 13

Availability

Imagine AJAX which is returning result. ...sometimes..

Sunday, 22 September 13

Partition Tolerance

Node A can see Node B always!

Sunday, 22 September 13

Partition Tolerance

If you have only Node A === Node A cant see Node A

Sunday, 22 September 13

BASE Jump

Sunday, 22 September 13

Basically Available Soft-State Eventually consistent

Sunday, 22 September 13

Highlights
Its impossible! YES IT IS! Focus on CA, AP, CP.

Sunday, 22 September 13

CAP Highlights

You are assuming that software/network/hardware failures will not happen

Sunday, 22 September 13

CAP Highlights

Your solution is equivalent to an existing one that doesn't beat CAP

Sunday, 22 September 13

CAP Highlights

Read-only mode is still unavailability for writes

Sunday, 22 September 13

CAP Highlights

Your system accumulates data forever and assumes innite storage

Sunday, 22 September 13

CAP Highlights

You are in charge of people's data

Sunday, 22 September 13

CAP Highlights

...always...

Sunday, 22 September 13

You might also like