is cache-only-prefixes an nnpfs limitation?

[Niels Möller]

Tomas Olsson <tol at stacken.kth.se> writes:

> Currently arlad takes care of all cache replacement things, on a node
> level. The problem is that it's really nnpfs that gets all the syscalls and
> knows what nodes are used, so arlad just somes up with nodes that it
> *thinks* are not recently used, and says "can these be dropped?".
> 
> For blocks (especially when we start talking about files larger than your
> cache) things get more complicated.  So we figured it's time for nnpfs to
> use the information available to it, and assume responsibility for the LRU.

One possibly naive idea:

1. Let arlad keep the responsibility for cache management.

2. Let nnpfs keep track of the 1000 most recent operations (or most
   recently accessed blocks, whatever). When the buffer with
   information about 1000 operations is full, dump that information on
   arlad and forget it.

The idea is that one call from nnpfs to arlad per 1000 file operations
should be a fairly small overhead, and then most of the complexity and
data structures can stay in userspace.

Regards,
/Niels