Hi everyone. One question is about performance of bucket algorithms, it seems that on every aspect, straw bucket is at least not worse than list bucket. So does it mean we shall always use straw bucket instead of list bucket?
I was browsing branch tree of ceph to trace back earlier versions of the implementation. I noticed that crush_uniform_bucket_choose has been significantly changed. I read the comment which said the original method (seems to be described in the thesis) is not random enough and has some bad behavior, but I don't know what kind of bad behavior is that. For what reason a random permutation is a better choice? Regards, Sam -- To unsubscribe from this list: send the line "unsubscribe ceph-devel" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html
