A caching model of operating system kernel functionality

摘要: Operating system research has endeavored to develop micro-kernels that provide modularity, reliability and security improvements over conventional monolithic kernels. However, the resulting kernels have been slower, larger more error-prone than desired. These efforts also failed sufficient application control of resource management required by sophisticated applications.This paper describes a caching model operating functionality as implemented in Cache Kernel, supervisor-mode component V++ system. The Kernel caches objects such threads address spaces just hardware memory data. User-mode handle loading writeback these objects, implementing application-specific policies mechanisms. Experience with measurements its performance on multiprocessor suggest can competitive systems, yet provides application-level resources, better scalability, smaller size basis for fault containment.