Although the processing power of uniprocessor systems is significantly increasing every year, there is a never ending quest for peak performance. The so called grand challenges in computing require more computing power than a single uniprocessor machine can ever offer. Recent advances in parallel processing however has brought the solution of grand challenges into reach. What is usually neglected when reporting on peak performance and Teraflops is the effort which would be necessary for implementing an efficient application. Handling all the communication, which is needed in a parallel algorithm, is probably not the most difficult part for the programmer, but often the most annoying part. To support or even automate this task a variety of programming tools and libraries has been developed. An interesting question is how to investigate and compare the performance obtained when using such tools and the effort for using these tools. For that purpose an evaluation project has been started on a cluster of nine workstations using the five most commonly used tools. The results are discussed.