Hierarchical Workload Characterization for a Busy Web Server
TOOLS '02 Proceedings of the 12th International Conference on Computer Performance Evaluation, Modelling Techniques and Tools
Measuring the Normality of Web Proxies' Behavior Based on Locality Principles
NPC '08 Proceedings of the IFIP International Conference on Network and Parallel Computing
PSnAP: accurate synthetic address streams through memory profiles
LCPC'09 Proceedings of the 22nd international conference on Languages and Compilers for Parallel Computing
| Hi-index | 4.10 |
To analyze or simulate an operating/computer system, one must construct a model of the programs executing within the system. Many recent analyses, particularly queueing models, have used mathematically convenient program models. For example, in one popular model the times between page faults are assumed to be exponentially distributed and independent. However, such a program model is inaccurate, which may cause the analysis to be unrepresentative of the real world. Simulation models of systems, on the other hand, have relied largely on traces of actual programs. Such traces are undoubtedly more accurate than simple mathematical models, but they have several drawbacks. They are expensive to generate, they may not be truly representative of typical programs, and they may contain more detail than is necessary for accurate system modeling. Moreover, it is difficult to extrapolate their behavior to other similar programs.