Designing multicore scalable filesystems with durability and crash consistency

Abstract

It is challenging to simultaneously achieve multicore scalability and high disk throughput in a file system. For example, data structures that are on separate cache lines in memory (e.g., directory entries) are grouped together in a transaction log when the file system writes them to disk. This grouping results in cache line conflicts, thereby limiting scalability. McoreFS is a novel file system design that decouples the in-memory file system from the on-disk file system using per-core operation logs. This design facilitates the use of highly concurrent data structures for the in-memory representation, which allows commutative operations to proceed without conflicts and hence scale perfectly. McoreFS logs operations in a per-core log so that it can delay propagating updates to the disk representation until an fsync. The fsync call merges the per-core logs and applies the operations to disk. McoreFS uses several techniques to perform the merge correctly while achieving good performance: timestamped linearization points to order updates without introducing cache line conflicts, absorption of logged operations, and dependency tracking across operations. Experiments with a prototype of McoreFS show that its implementation is conflict-free for 99% of test cases involving commutative operations generated by Commuter, scales well on an 80-core machine, and provides disk performance that matches or exceeds that of Linux ext4.