Skip to content
Open access

PRDCC: PRedictive Deterministic Concurrency Control under High Contention

Sep 2026 · Proceedings of the ACM on Management of Data · Vol 4, pp. 1-26 · 0 citations · 19 references

Abstract

Deterministic databases present a promising direction for building scalable OLTP systems by establishing a global transaction order prior to execution, thereby eliminating costly runtime coordination overhead. However, existing methods face significant limitations: the Ordering-Constrained protocols strictly commit transactions in a predetermined TID order, limiting concurrency and scalability; meanwhile, execution-order-independent approaches can incur substantial conflicts, cascading aborts, and retries under high-contention workloads. In this paper, we propose PRDCC, a prediction-driven framework for deterministic concurrency control, designed to mitigate abort-induced throughput collapse under high data contention. PRDCC combines (i) a deterministic conflict predictor for extractable point/range predicates that avoids physical pre-execution and provides a zero-false-negative guarantee within its supported predicate class, and (ii) a deterministic partitioning algorithm that uses predicted dependencies to form an ordered sequence of execution blocks. Experimental results show that PRDCC improves throughput by up to 4.1× over Aria on YCSB and 3.06× on TPC-C under high contention, while maintaining competitive performance in low-contention workloads.

Read PDF

Similar papers

Open access Oct 2026

Efficient Predictive Monitoring of Message Passing Interface Programs

The Message Passing Interface (MPI) is the standard programming model for high-performance computing, yet nondeterministic scheduling in concurrent executions makes reliability assurance highly challenging. Beyond deadlocks, property-related bugs such as modifying a send buffer before a nonblocking send completes or ac...

Jia-Qiang Yao, Hao-Cheng Geng, Zhen-Bang Chen · 0 citations

DynBranch: Speculative Subgraph Reuse for Dynamic Agentic LLM Serving

DynBranch is proposed, which makes an unresolved branch addressable before it resolves, and its stable coordinate lets candidate subgraphs run during resolution and completed subgraph results be reused across later requests.

Jun-Yi Shen, Noppanat Wadlom, Zheng-Yuan Su et al. · 0 citations
Preprint Oct 2026

TenonAgent: Virtualizing Execution Residency for Memory-Oversubscribed LLM Agents

LLM agents execute as long-lived, multi-stage workflows with substantial resource-demand variation across phases. Provisioning each agent or sandbox for its lifetime peak leads to low resource utilization, particularly for memory, whereas aggressive oversubscription risks memory contention, out-of-memory failures, and...

Yi-Fan Zhang, Xin-Kui Zhao, Rui-Xin Huang et al. · 0 citations
Preprint Nov 2026

Optimizing Geo-replicated Deterministic Databases

In transactional databases, achieving active-active replication among geo-distributed replicas is challenging due to the high coordination cost in ensuring the serializability of concurrent cross-node transactions. Deterministic concurrency control (DCC) has emerged as a promising alternative: it establishes a global o...

Yun-Hao Mao, Shi-Quan Zhang, Geng-Rui Zhang et al. · 0 citations
#natural language process... Preprint Oct 2026

Flide: Accelerating Decision Inference with Exact State Forking

Decision models such as Jev have emerged as a highly promising paradigm for LLM agent systems, providing lightweight and reliable categorical reasoning for rapid System One execution. However, existing generation-oriented inference engines fail to support these workloads effectively, suffering from severe prefix recomp...

Bobo Li, Jia-Jia Song, Hao Fei et al. · 1 citation · ⚡1
Preprint Nov 2026

Qubrio: High-Performance Quantum Compilation via Multi-Agent LLM Collaboration

Large Language Models (LLMs) offer an adaptable alternative to traditional quantum compilation heuristics, which bottleneck progress by demanding costly manual redesigns whenever hardware evolves. However, naive single-pass LLM approaches fail under the unique challenges introduced by quantum compilation. We introduce...

Ji-Xuan Ruan, Zhuo Cui, Zheng-Ding Hu et al. · 0 citations

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.