Comment #2 on issue 374 by [email protected]: Add method in ByteString to avoid big buffer copy.
http://code.google.com/p/protobuf/issues/detail?id=374
In Java, copying tends to represent a much smaller chunk of execution time. I'm surprised this actually makes that big a difference unless you have really large byte/message fields, which protocol buffers aren't really meant to address.
Particularly because of their immutability (and of course garbage collection), my Message objects tend only be copying during IO. In Java, you *can* currently reduce some copying during IO through the use of unbuffered steams and memory mapped files. It can be a big win, though often is not. Have you tried doing this and seen a big win?
-- You received this message because you are subscribed to the Google Groups "Protocol Buffers" group. To post to this group, send email to [email protected]. To unsubscribe from this group, send email to [email protected]. For more options, visit this group at http://groups.google.com/group/protobuf?hl=en.
