Distributed Systems

← Back to document page

Distributed SystemsUniversity of CambridgeComputer Science Tripos, Part IBMichaelmas term 2021/22Course web page: videos: Martin About Distributed Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Distributed Systems and computer networking . . . . . . . . . . . . . . . . . . . . . . . . . Example: Remote Procedure Calls (RPC) . . . . . . . . . . . . . . . . . . . . . . . . . . .82 Models of Distributed The two generals problem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . The Byzantine generals problem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Describing nodes and network behaviour . . . . . . . . . . . . . . . . . . . . . . . . . . . . fault tolerance and high availability.

If one component of a system stops working, we call that a fault, and many distributed systems strive to provide fault tolerance: that is, the system as a whole continues functioning despite the fault. Dealing with faults is what makes distributed computing fundamentally di erent, and often harder, compared to programming a single computer.

  Fault, Tolerance, Fault tolerance