Abstract:Digital signatures are fundamental building blocks in various protocols to provide integrity and authenticity. The development of the quantum computing has raised concerns about the security guarantees afforded by classical signature schemes. CRYSTALS-Dilithium is an efficient post-quantum digital signature scheme based on lattice cryptography and has been selected as the primary algorithm for standardization by the National Institute of Standards and Technology. In this work, we present a high-throughput GPU implementation of Dilithium. For individual operations, we employ a range of computational and memory optimizations to overcome sequential constraints, reduce memory usage and IO latency, address bank conflicts, and mitigate pipeline stalls. This results in high and balanced compute throughput and memory throughput for each operation. In terms of concurrent task processing, we leverage task-level batching to fully utilize parallelism and implement a memory pool mechanism for rapid memory access. Considering the impact of varying repetition numbers in Dilithium on the overall execution time and hardware utilization, we propose a dynamic task scheduling mechanism to improve multiprocessor occupancy and significantly reduce execution time. Furthermore, we apply asynchronous computing and launch multiple streams to hide data transfer latencies and maximize the computing capabilities of both CPU and GPU. Across all three security levels, our GPU implementation can concurrently compute ten thousand tasks in less than 32 miliseconds for signing and 15 miliseconds for verification on both commercial and server-grade GPUs. This achieves microsecond-level amortized execution times for each task, offering a high-throughput and quantum-resistant solution suitable for a wide array of applications in real systems.

High-Speed Privacy Amplification Scheme Using GMP in Quantum Key Distribution

High-Speed Implementation of Length-Compatible Privacy Amplification in Continuous-Variable Quantum Key Distribution

Large-Scale and High-Speed FPGA-Based Privacy Amplification for Quantum Key Distribution

High-Performance Hardware Implementation of the Saber Key Encapsulation Protocol

Designing a Scalable and Area-Efficient Hardware Accelerator Supporting Multiple PQC Schemes

PQPU: A 4.4- $\mu$ J/Op 69.4-Kops Agile Post-Quantum Crypto-Processor Across Multiple Mathematical Problems

Practical, high-speed Gaussian coherent state continuous variable quantum key distribution with real-time parameter monitoring, optimised slicing, and post-processed key distillation

Efficient Parallelism of Post-Quantum Signature Scheme SPHINCS

HMNTT: A Highly Efficient MDC-NTT Architecture for Privacy-preserving Applications

High-Speed Variable Polynomial Toeplitz Hash Algorithm Based on FPGA

A 334$μ$W 0.158mm$^2$ ASIC for Post-Quantum Key-Encapsulation Mechanism Saber with Low-latency Striding Toom-Cook Multiplication Authors Version

Overcoming Noise Limitations in QKD with Quantum Privacy Amplification

Quantum Privacy Amplification for A Sequence of Single Qubits

Hardware Design and Optimization of Polynomial Multiplication for Post-Quantum Cryptography Algorithm Based on NTT

A Number Theoretic Transform Accelerator with Two Parallel Simplified Butterfly Units

Overcoming noise limitations in quantum key distribution with quantum privacy amplification

Towards High-Performance Supersingular Isogeny Cryptographic Hardware Accelerator Design

100Mbps Reconciliation for Quantum Key Distribution Using a Single Graphics Processing Unit

High-throughput GPU layered decoder of quasi-cyclic multi-edge type low density parity check codes in continuous-variable quantum key distribution systems

16.2 A 28nm 69.4kOPS 4.4μJ/Op Versatile Post-Quantum Crypto-Processor Across Multiple Mathematical Problems.

High-Throughput GPU Implementation of Dilithium Post-Quantum Digital Signature