Quite often one has to manage a set of computers which are used by many others. A typical example is a computer lab.
In such a case, not all computers may be up at a given time. This can be a headache for system administrators -- they must turn all of the computers on to run updates and cleanup, or they must update the machines incrementally.
It is quite possible to run a script that continually tries to run a given command on all the machines until it succeeds. However, this requires a "leader" machine that is guaranteed to be on for a long period of time. Such a machine is not always available.
Dash solves this problem by using Raft to replicate commands across machines. Simply calling
dash apt-get update1 from a machine on the same network will instruct the machines to replicate the command across the cluster,
and run it when it has been replicated to a majority. If a machine is booted up later, it will be told by the other machines to run the command too.
This was made as a project for CS 733 (Advanced Cloud Computing) at IIT Bombay, taught by Professor Sriram Srinivasan
1. Currently Dash doesn't have any security model so for this to work it will need to be run as root and will be vulnerable. ECDSA-based security is planned.