Jay Lepreau, Mike Hibler, Bryan Ford, Jeffrey Law, and Douglas Orr. In-Kernel Servers on Mach 3.0: Implementation and Performance. Proceedings of the Third USENIX Mach Symposium, April, 1993, pp. 39-56. Home/Search Document Not in Database Summary Related Articles Check |
This paper is cited in the following contexts:
The Impact of Operating System Structure on Memory System.. - Chen, Bershad (1993) (97 citations) (Correct)
....for a system instructions executed. The last category, S MCPI, small portion of overall system overhead. This suggests reflects relative memory system overhead for system that microkernel optimizations focusing exclusively on references. Four of the activities (Microkernel, Emulator, IPC [8, 18, 20, 26, 34], without considering other sources IPC, VM mi) occur only in Mach. Block Ops for Mach of system overhead such as MCPI, will have a limited includes operations from both the Mach kernel and the impact on overall system performance [7] UNIX Server. The number at the top of each bar is system ....
Jay Lepreau, Mike Hibler, Bryan Ford, Jeffrey Law, and Douglas Orr. In-Kernel Servers on Mach 3.0: Implementation and Performance. Proceedings of the Third USENIX Mach Symposium, April, 1993, pp. 39-56.
The Impact of Software Structure and Policy on CPU and Memory.. - Chen (1994) (5 citations) (Correct)
....builders need to learn more about partitioning systems and how it affects performance. Towards this end, several recent research projects propose new ways of partitioning systems implementations to achieve some of the engineering advantages of microkernel structure without sacrificing performance [21, 50, 51, 83]. It also means that the implementors of microkernel systems must give particular attention to performance issues. System implementors have typically used elapsed time or throughput to measure and compare system performance. This thesis has demonstrated that address traces and trace driven ....
....the operation. Simple tools such as timing devices will always be important, but for identifying and understanding the complex interactions that occur in a multiple address space system these simple tools are not always good enough. Variations on the partitioning of Mach 3. 0 have been proposed [21, 50, 51]. Figure 4 2 shows that if such changes are to be effective, they must reduce the dynamic instruction count for the system. Some of the behavior identified in Figure 4 2 can be attributed to specific parts of the system. For example, the higher overhead of VM activity in Mach 3.0 is due to the ....
Jay Lepreau, Mike Hibler, Bryan Ford, Jeffrey Law, and Douglas Orr. In-Kernel Servers on Mach 3.0: Implementation and Performance. Proceedings of the Third USENIX Mach Symposium, April, 1993, pp. 39-56.
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC