On our Fuse doc:
Most of the overheads come from the fact that there are several memory copies going on for each call on `read` or `write` operations, and that FUSE caps the maximum granularity of writes to 128KB. This could be probably improved by a large extent by leveraging the FUSE cache write-backs feature introduced in kernel 3.15 (not supported yet, however, by libfuse 2.x userspace libs).
Now the Fuse writeback_cache is supported in the Fuse 3.X. This enables write-back caching which can significantly improve performance.
Investigate and see if there are any performance improvement.