Evaluate Direct I/O for better performance

[Bukhtawar]

Is your feature request related to a problem? Please describe.
The primary benefit of direct I/O is to reduce CPU utilization for file reads and writes by eliminating the copy from the cache to the user buffer. Many operations like peer recovery, segrep transferring large merged segments to replica for instance w/o direct I/O can cause other segments(since refreshes no longer leverage shard idle #7736 ), being queried heavily to be pushed out of the cache even if. The remote store interactions needs to be revisited to explore any opportunities for improvements

Describe the solution you'd like
A clear and concise description of what you want to happen.

Describe alternatives you've considered
A clear and concise description of any alternative solutions or features you've considered.

Additional context
Add any other context or screenshots about the feature request here.

[msproact]

Another approach here would be to use posiv_fadvice() with POSIX_FADV_DONTNEED (see manpage posix_fadvise(2))¹. However I bet Direct I/O could even be more efficient.