Coursera Flash Sale
40% Off Coursera Plus for 3 Months!
Grab it
Watch this 14-minute conference presentation from OOPSLA 2025 that introduces a novel approach to implementing and verifying strongly-consistent replicated data stores using weak replicated data stores as a foundation. Learn how researchers Nicholas V. Lewchenko, Gowtham Kaki, and Bor-Yuh Evan Chang from the University of Colorado Boulder address the complexity challenges in implementing consensus algorithms like Raft and Paxos by abstracting away asynchronous communication details. Discover the Super-V verified-programming framework that enables SMT-based verification through local-scope stable update preconditions rather than traditional global inductive invariants, making automated verification more accessible to developers. Explore how this methodology successfully implements and verifies a strong replication system based on an adapted Raft consensus algorithm, offering a more manageable approach to building fault-tolerant distributed systems while maintaining strong consistency guarantees under network partitions.
