Achieved IPC performance (still the foundation for extensibility)

Extensibility can be based on cross-address-space interprocess communication (IPC) or on grafting application-specific modules into the operating system. For comparing both approaches, we need to explore the best achievable performance for both models. This paper reports the achieved performance of cross-address-space communication for the L4 microkernel on Intel Pentium, Mips R4600 and DEC Alpha processors. The direct costs range from 45 cycles (Alpha) to 121 cycles (Pentium). Since only 2.3% of the L1 cache are required (Pentium), the average indirect costs are not to be expected much higher.