[Paper Review] OPTIMUM-DERAM: Highly Consistent, Scalable, and Secure Multi-Object Memory using RLNC

Nicolas Nicolaou, Kishori M. Konwar+4 more|arXiv (Cornell University)|Jan 19, 2026

Distributed systems and fault tolerance0 citations

TL;DR

Optimum-DeRAM provides a decentralized, atomic read/write memory supporting multiple objects, using RLNC to improve scalability, storage efficiency, and Byzantine tolerance, with dynamic join/leave via a registry oracle.

ABSTRACT

This paper introduces OPTIMUM-DERAM, a highly consistent, scalable, secure, and decentralized shared memory solution. Traditional distributed shared memory implementations offer multi-object support by multi-threading a single object memory instance over the same set of data hosts. While theoretically sound, the amount of resources required made such solutions prohibitively expensive in practical systems. OPTIMUM-DERAM proposes a decentralized, reconfigurable, atomic read/write shared memory (DeRAM) that: (i) achieves improved performance and storage scalability by leveraging Random Linear Network Codes (RLNC); (ii) scales in the number of supported atomic objects by introducing a new object placement and discovery approach based on a consistent hashing ring; (iii) scales in the number of participants by allowing dynamic joins and departures leveraging a blockchain oracle to serve as a registry service; and (iv) is secure against malicious behavior by tolerating Byzantine failures. Experimental results over a globally distributed set of nodes, help us realize the performance and scalability gains of OPTIMUM-DERAM over previous distributed shared memory solutions (i.e., the ABD algorithm [3])

Motivation & Objective

• Develop a scalable, atomic read/write shared memory by composing multiple objects (multi-object memory).
• Leverage Random Linear Network Coding (RLNC) to improve fault tolerance, storage efficiency, and latency.
• Enable dynamic joins/departures and scalable object placement without performance degradation.
• Achieve Byzntine fault tolerance through quorum-based protocols and signatures.
• Provide a reusable framework that scales with object size, object count, node count, and operation concurrency.

Proposed method

• Use RLNC to encode object values into coded elements where any k codes suffice to decode.
• Introduce a consistent hashing ring to distribute objects across node clusters for scalable placement.
• Employ a blockchain/consensus-based oracle as a registry to support dynamic joins/departures of nodes.
• Adopt a quorum-based protocol with signed tags and coded elements to tolerate Byzantine behaviors.
• Define data access primitives (DAPs) and prove safety (atomicity) and liveness under Byzantine faults.
• Evaluate the protocol against the ABD baseline focusing on object size, number of objects, node count, and concurrency.

Experimental results

Research questions

• RQ1 How can atomic read/write memory be implemented for multiple objects using RLNC in a decentralized setting?
• RQ2 How can object placement and discovery be scaled across clusters with a consistent hashing ring?
• RQ3 How can dynamic joins/departs be supported without disrupting concurrent operations?
• RQ4 What level of Byzantine fault tolerance is achievable with RLNC-based storage and signed elements?
• RQ5 How does Optimum-DeRAM perform compared to traditional ABD in terms of scalability across core dimensions?

Key findings

• Optimum-DeRAM demonstrates scalability improvements over classic ABD in experiments.
• The approach focuses on four scalability dimensions: object size, number of objects, number of nodes, and operation concurrency.
• RLNC encoding reduces storage and communication overhead while enabling recovery from a subset of coded elements.
• A consistent hashing ring enables scalable multi-object distribution across clusters.
• Join/leave protocols with a registry service allow dynamic participation without compromising safety.
• The protocol provides Byzantine tolerance through quorum intersections and signed coded elements.