@seiflotfy Thanks for publicizing PMC. I'm very interested in the benchmarking blogpost you mentioned on HN that compares CMS, CML, PMC etc.
@seiflotfy ah so DPC seems a bit like vHLL - really it's for counting distinct elements per flow.
-
-
@avibryant It focuses on the flow sizes ==> makes it cardinality oriented. If sizes are random then results should be similar to PMC. Right? -
@seiflotfy instead of random numbers they use a hash of the header + first 12 bytes of the content of packet. -
@seiflotfy so if you see the same packet two different places, then later merge the sketches, it only gets counted once (vs twice with PMC). -
@avibryant haven't completed reading the paper yet. But that sounds very right :D On another note I am confused on why no1 picked up PMC yet -
@avibryant The paper came out in 2011 -
@seiflotfy yeah very surprising, given how well-known HLL and CMS have become. I'm looking forward to doing PMC in Scala when I get time. -
@avibryant Btw are you familiar with#CuckooFilter https://github.com/seiflotfy/cuckoofilter … ? -
@seiflotfy yes, I just don't care about deletions that much :) - 1 more reply
New conversation -
-
-
@avibryant I am thinking if it is actually possible to get the grand cardinality across all flows via DPC? Any thought?Thanks. Twitter will use this to make your timeline better. UndoUndo
-
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.