Lecture Distributed and Parallel Programming

Description

VAK 04IN2052 / 432052
Category: Lecture+Lesson, 4 SWS
Master Course
ECTS: 6, Sommer Semester
University of Koblenz-Landau and University of Bremen Lecturer: PD Dr. Stefan Bosse

Learning content

Introduction of multiprocess models with process algebra and process flow diagrams based on Hoare's CSP model
Expansion of the CSP model with competition and shared resources; conflict resolution
Introduction of parallels and distributed programming languages (OCCAM, Functional Programming, Multi-threading, MPI, JavaScript)
Synchronization and inter-process communication; Errors like deadlocks and starvation; Properties of parallel and distributed systems
Synchronization Objects and their application in Programming {Mutex, Semaphore, Monitor, Event, Queue, Barrier, Channel}
Computer Architectures for Parallel and Distributed Systems (MultiCore, GPU, Cluster)
Possibilities of parallelization and distribution of common algorithms and programs
Formal foundations of parallel systems; metrics; efficiency; scaling
Election, Leader, consensus algorithms and group communication

Goals / competences

The students acquire / win / learn

Understanding the basic principles and architectures of distributed (VS) and parallel systems (PS) and the ability to transfer to technical systems
Understanding and ability of programmatic application of synchronization and communication in VS and PS
Understanding the problems and operation of parallel systems versus sequential systems (efficiency, blocking, scaling, resource requirements)
Practical knowledge of programming PS and VS using programming exercises with JavaScript and node.js
Insights into the limits and possibilities of parallelization and distribution and the ability to develop efficient systems

Download

Script: PDF
Slides: PDF