Sark: Oblivious Integrity Without Global State

TL;DR

Sark addresses the problem of providing oblivious, non-custodial asset management with remote integrity without relying on a global, fully visible ledger. It achieves this via a modular architecture comprising Client, Porter, and Validator subsystems, and the USO asset model that enables private, verifiable state updates anchored to a blockchain. The work analyzes the CIA trade-offs, introduces the Integrity Locus concept, and evaluates performance with Raft-based finality and different Merkle data structures, outlining future work toward BFT, Porter clustering, and availability enhancements. The proposed approach offers robustness to governance topology and data confidentiality at the cost of data availability and local centrality, with potential applicability to non-monetary digital assets like diplomas or identity proofs in regulated environments.

Abstract

In this paper, we introduce Sark, a reference architecture implementing the Unforgeable, Stateful, and Oblivious (USO) asset system as described by Goodell, Toliver, and Nakib. We describe the motivation, design, and implementation of Sloop, a permissioned, crash fault-tolerant (CFT) blockchain that forms a subsystem of Sark, and the other core subsystems, Porters, which accumulate and roll-up commitments from Clients. We analyse the operation of the system using the 'CIA Triad': Confidentiality, Availability, and Integrity. We then introduce the concept of Integrity Locus and use it to address design trade-offs related to decentralization. Finally, we point to future work on Byzantine fault-tolerance (BFT), and mitigating the local centrality of Porters.

Figures (11)

• Figure 1: Sark Architecture
• Figure 2: Sark Vertical Implementation
• Figure 3: Sark USO Asset
• Figure 4: Alice registers an update, giving Bob control first and possession later.
• Figure 5: Sark Protocol Diagram