Memory Segment efficient array handling

[Maurizio Cimadamore]

On 01/04/2021 16:32, Uwe Schindler wrote:
> New code: Map the huge file into few MappedMemorySegments (not too large to prevent fragmentation issues). Each memory segment sliced using MemorySegment.asSlice(...).asByteBuffer(). Access speed should be identical to current code, just less mappings in kernel needed and we have safe unmapping.

Yeah - that should work. Note that, right now shared segments and 
buffers have an issue when it comes to async IO. This issue will be 
resolved now that we have the new ResourceScope API, but the code which 
fixes this is not there yet. So depending on whether you use these or 
not, you might see some issues.

That said, we should really look into where the performance issues come 
from... I know there's a bunch of workarounds in place for the long vs. 
int (which might go away soon), but historically these workarounds have 
served us well; it would be interesting here to understand if the issues 
you are seeing are "more of the same" or a new issue.

Maurizio