PROFET: Modeling System Performance and Energy Without Simulating the CPU

Application performance on novel memory systems is typically estimated using a hardware simulator. The simulation is, however, time consuming, which limits the number of design options that can be explored within a practical length of time. Also, although memory simulators are typicallywell validated, current CPU simulators have various shortcomings, such as simplified out-of-order execution, an obsolete data prefetcher and a lack of virtual-to-physical memory translation, all of which can make a huge difference between the simulated and actual memory system.