Skip to content

suranap/scarecrow

Folders and files

NameName
Last commit message
Last commit date

Latest commit

Jul 14, 2011
ecff310 · Jul 14, 2011

History

4 Commits
 
 
 
 

Repository files navigation

Introduction

Scarecrow uses Prolog to monitor and administer resources on Amazon Web Services. It is (or will be) an expert system for manipulating AWS automatically.

The current state of your AWS account is called the initial state (I). The state you wish to achieve is called the goal state (G). The goal state could be a web deployment, a DB backup, a modified security group, etc. How do you get from I to G? You could use the AWS management console to manually perform the required actions, but this is tedious and error-prone. You could instead write a script to perform these same actions programmatically, but you would need to write different scripts for every different task or goal.

Scarecrow, by contrast, is sort of like a script generator. All you need to do is declare the goal state you wish to reach. It will query AWS to discover the initial state. Then it will use a library of predefined domain-specific rules to determine the sequence of actions required to get from I to G.

Why Prolog? The language is unique for its powerful support for backtracking and unification. The first is used to search for a path from I to G to accomplish a task. The latter is used to make complex queries on the state of your AWS deployment. Think of it as a weird SQL query.

How is this different from boto or a million other libraries for AWS? These libraries are merely a programming interface to AWS, but you still have to write all the scripts yourself. Scarecrow is a level up from this: it writes the scripts for you. Well, it might someday...

Here's another attempt at explaining the purpose of Scarecrow.

Requirements

About

Using Prolog to administer AWS deployments

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published