Learn More
With the advent of 64-bit processors, virtual address spaces will be large enough to make it feasible for a distributed operating system to support as the primary system abstraction the illusion of a uniform shared address space that spans a collection of workstations. Under such a system, interprocess communication can be performed by invoking functions(More)
Modern operating systems must support a wide variety of services for a diverse set of users. Designers of these systems face a tradeoo between functionality and performance. Systems like Mach provide a set of general abstractions and attempt to handle every situation, which can lead to poor performance for common cases. Other systems, such as Unix, provide(More)
For a parallel architecture to scale eeectively, communication latency between processors must be avoided. We have found that the source of a large number of avoidable cache misses is the use of hardwired write-invalidate coherency protocols, which often exhibit high cache miss rates due to excessive invalidations and subsequent reloading of shared data. In(More)
As the gap between processor and memory speeds widens, system designers will inevitably incorporate increasingly deep memory hierarchies to maintain the balance between processor and memory system performance. At the same time, most communication subsystems are permitted access only to main memory and not a processor's top level cache. As memory latencies(More)
  • 1