I've spent some time investigating memory mapped IO for an application I'm working on. I have some very large (TB scale) files, and I want to map segments from them into memory, for both reading and writing, making maximum use of OS-level caching. The software I'm writing needs to work under Unix/Linux and Windows... performance is critical.
I've discovered boost::iostreams::mapped_file_source
and boost::iostreams::mapped_file_sink
, which provide most of the facilities I'm looking for. The facilities I'd like, but haven't found are:
- Forcing a synchronisation of written data to disk (
msync
(2) on Unix;FlushViewOfFile
on Windows) - Locking of files to prevent two processes attempting to write the same file at the same time (or read while the file is still being written..)
- Controlling attributes of the file at creation time (Unix)
Can I do these things using "boost/iostreams/device/mapped_file.hpp"
? Are there other platform independent libraries that would better suit my requirements? Must I develop my own cross-platform library to get this flexibility?