A TimeStamp based Multi-version STM Protocol that satisfies Opacity and Multi-Version Permissiveness
Priyanka Kumar, Sathya Peri

TL;DR
This paper introduces a new timestamp-based multi-version STM protocol that guarantees opacity and mv-permissiveness, with formal proofs of correctness and a garbage collection method.
Contribution
It presents the first formally proven multi-version STM system satisfying both opacity and mv-permissiveness, including a verified garbage collection process.
Findings
The proposed STM system guarantees opacity.
The system is mv-permissive, avoiding aborts for read-only transactions.
Formal proofs validate the correctness of the protocol and garbage collection.
Abstract
Software Transactional Memory Systems (STM) are a promising alternative to lock based systems for concurrency control in shared memory systems. In multiversion STM systems, each write on a transaction object produces a new version of that object. The advantage obtained by storing multiple versions is that one can ensure that read operations do not fail. Opacity is a commonly used correctness criterion for STM systems. Multi-Version permissive STM system never aborts a read-only transaction. Although many multi-version STM systems have been proposed, to the best of our knowledge none of them have been formally proved to satisfy opacity. In this paper we present a time-stamp based multiversion STM system that satisfies opacity and mv-permissiveness. We formally prove the correctness of the proposed STM system. We also present garbage collection procedure which deletes unwanted versions of…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsDistributed systems and fault tolerance · Cognitive Functions and Memory · Real-Time Systems Scheduling
