Ralph et al, > I'm also particullary curious of trying some alternative, high > performance and thread conscious malloc like tcmalloc and horde with > Judy. I'll let you know how this performs in my use scenario.
Sure, sounds interesting. We originally built Judy including its own internal "small block allocator" essentially to minimize calls to malloc() and free(). Much later after many experiments, Doug concluded that was silly and ripped it out, saying the right answer is to link with the best malloc() you can find. At the time that was Doug Lea (spelling?) open-source malloc(). > Just out of curiosity, have you ever tought about an integration of a > fast, light form of locking (futex based, i.e.) with Judy? Guess not, since I don't know what futex is. You must realize that our small team had our hands full just producing Judy IV (the fourth and final version, massive rewrite, the one put on SourceForge) before HP cancelled the project. We only scratched the surface in several directions, such as shipping Judy with it's own "good malloc()" (but then the linking application must use it too), studying locking issues, writing formal academic papers, etc. Alan ------------------------------------------------------------------------- Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT & business topics through brief surveys-and earn cash http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV _______________________________________________ Judy-devel mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/judy-devel
