This preview has intentionally blurred sections. Sign up to view the full version.View Full Document
Unformatted text preview: On p-Kernel Construction Abstract Jochen Liedtke GMD — German National Research Center for Information Technology * [email protected] .de From a software-technology point of view, the p-kernel concept is superior to large integrated kernels. On the other hand, it is widely believed that (a) p-kernel based systems are inherently inefficient and (b) they are not sufficiently flexible. Contradictory to this belief, we show and support by documentary evidence that inef- ficiency and inflexibility of current p-kernels is not in- herited from the basic idea but mostly from overloading the kernel and/or from improper implementation. Based on functional reasons, we describe some con- cepts which must be implemented by a p-kernel and illustrate their flexibility. Then, we analyze the per- formance critical points. We show what performance is achievable, that the efficiency is sufficient with re- spect to macro-kernels and why some published contra- dictory measurements are not evident. Furthermore, we describe some implementation techniques and illustrate why p-kernels are inherently not port able, although they improve portability of the whole system. 1 Rationale p-kernel based systems have been built long before the term itself was introduced, e.g. by Brinch Hansen  and Wulf et al. . Traditionally, the word ‘kernel’ is used to denote the part of the operating system that is mandatory and common to all other software. The basic idea of the ~-kernel approach is to minimize this part, i.e. to implement outside the kernel whatever possible. *GMD SET–RS, 53754 Sankt Augustin, Germany Permission to make digital/hard copy of part or all of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage, the copyright notice, the title of the publication and its date appear, and notice is given that copying is by permission of ACM, inc. To copy otherwise, to republish, to post on servers, or to redistribute to lists, requires prior specific permission and/or a fee. SIGOPS ’95 12/95 CO, USA 01995 ACM 0-89791 -715-419510012 ...$3.50 The software technological advantages of this ap- proach are obvious: (a) (b) (c) A clear p-kernel interface enforces a more modular system structure. 1 Servers can use the mechanisms provided by the ~-kernel like any other user program. Server mal- function is as isolated as any other user program’s malfunction. The system is more flexible and tailorable. Differ- ent strategies and APIs, implemented by different servers, can coexist in the system. Although much effort haa been invested in p-kernel construction, the approach is not (yet) generally ac- cepted. This is due to the fact that most existing p- kernels do not perform sufficiently well. Lack of effi- ciency also heavily restricts flexibility, since important mechanisms and principles cannot be used in practice due to poor performance. In some cases, the p-kernel interface...
View Full Document
- Fall '08
- Virtual memory, CPU cache, Address space, Pentium