The Linux kernel utilizes caches to speed up accesses to the high-usage objects that are normally stored on disks, such as data cache pages for files, inodes, and directory entries. These cache data types are implemented using separate chaining. In our experience, we found that some of the hash functions used in th circa 2000 Linux kernel can results, for certain access patterns, in very skewed data distributions over the chains. Better hash functions are desirable. In this paper, we analyze the data distribution and run-time efficiencies of a set of hash function for Linux inode and directory caches. Our analysis utilizes data traces from both an industry standard benchmark and two real world computing environments. Although some of the results were anticipated by Lever, who has carried out a similar study, we show that his results can be further improved.
The authors of these documents have submitted their reports to this technical report series for the purpose of non-commercial dissemination of scientific work. The reports are copyrighted by the authors, and their existence in electronic format does not imply that the authors have relinquished any rights. You may copy a report for scholarly, non-commercial purposes, such as research or instruction, provided that you agree to respect the author's copyright. For information concerning the use of this document for other than research or instructional purposes, contact the authors. Other information concerning this technical report series can be obtained from the Computer Science and Engineering Department at the University of California at San Diego, firstname.lastname@example.org.
[ Search ]