Since CDN simulations are known to be highly memory-intensive, in this paper, we argue the need for reducing the memory requirements of such simulations. We propose a novel memory-efficient data structure that stores cache state for a small subset of popular objects accurately and uses approximations for storing the state for the remaining objects. Since popular objects receive a large fraction of the requests while less frequently accessed objects consume much of the memory space, this approach yields large memory savings and reduces errors. We use bloom filters to store approximate state and show that careful choice of parameters can substantially reduce the probability of errors due to approximations.

The port-related business, though as old as the human society, is very vibrant and relevant to our daily lives in Singapore; it also offers many exciting challenges and opportunities. One of the primary challenges is to maintain the high-level of services that the Port of Singapore has been reputed for, and hence allow us to get ready for the next millennium. To achieve this goal, many aspects of the port and terminal operations must be carefully researched for continual improvements. The computer and information technologies can offer tremendous advantages in every aspect of the planning, simulation and decision supports. This project is a point in case, where the container handling process is simulated in details. To manage the complexities involved and obtain timely results, the simulations are carried out on an SGI parallel computer. Numerous attempts to achieve a better performance have been made through improvements in modeling, simulation engine, the underlying runtime system, ...

In this report, the paradigms for simulating the performance of parallel hardware and software are investigated and the principles of discrete event and parallel and distributed simulation are explained in this context.