Repository navigation
Performance issue with skip #73
Description
I'm using the lazy interface of Data.Binary.Get to parse large binaries from disk, where most parts are skipped on the first pass. When using skip directly, I get a severe performance and space usage problem when the skipped byte count is large (many megabytes). I'm not an expert on lazy bytestring internals, but it seems like the input data is held on for too long before being skipped ("PINNED" memory usage in -hc heap profile). Using this wrapper around skip makes it several orders of magnitude faster and does not explode on the heap (I'm using GHC 7.10.1):
import qualified Data.Binary.Get as G
skipMany :: Int -> G.Get ()
skipMany bytes =
replicateM_ rep (G.skip cs) >> G.skip rest
where
cs = 1024
(rep, rest) = bytes `quotRem` csMetadata
Metadata
Assignees
Labels
No labels
Activity
kolmodin commentedon May 11, 2015
Right. The current implementation of
skipconcatenates the bytes to be skipped, and then throws it away.This is unnecessarily expensive. I'll work on a more efficient implementation.
kolmodin commentedon May 11, 2015
I might add that even with the more efficient implementation, it will still force to read all the data from disk (or wherever it comes from) which might be inefficient for your case.
The most efficient implementation would be to
seekin the file descriptor and only read the parts needed.kolmodin commentedon May 30, 2015
Fixed in e46b7b6.
Update binary submodule to 0.7.5.0 release
Update binary submodule to 0.7.5.0 release