Squire: A General-Purpose Accelerator to Exploit Fine-Grain Parallelism on Dependency-Bound Kernels

2025-10-23

AI Summary

Data-dependent intensive kernels suffer from low parallel efficiency on general-purpose accelerators, while FPGAs and ASICs lack programmability. To address this trade-off, this paper proposes Squire, a novel general-purpose accelerator architecture. Its core innovation integrates lightweight in-order CPU cores with domain-specific acceleration units in a tightly coupled manner, augmented by an L2 direct-access mechanism and a low-latency on-chip interconnect network; it further extends the Neoverse-N1 microarchitecture to support dynamic programming workloads. Squire deploys per-core acceleration transparently—requiring no modifications to the software stack—thereby preserving generality while enabling fine-grained parallelism exploitation. Evaluated on dynamic programming kernels, Squire achieves 7.64× kernel-level speedup, 3.66× end-to-end application acceleration, 56% energy reduction, and incurs only 10.5% additional hardware area overhead.

Technology Category

• Machine Learning: Hardware-aware ML
• Data Mining & Knowledge Management: Intelligent Query Processing
• Search and Optimization: Learning to Search

Application Category

• Search and Retrieval-Augmented AI: Vertical and domain-specific search
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Data management and stream processing for Web, mobile and wireless applications

Abstract

Multiple HPC applications are often bottlenecked by compute-intensive kernels implementing complex dependency patterns (data-dependency bound). Traditional general-purpose accelerators struggle to effectively exploit fine-grain parallelism due to limitations in implementing convoluted data-dependency patterns (like SIMD) and overheads due to synchronization and data transfers (like GPGPUs). In contrast, custom FPGA and ASIC designs offer improved performance and energy efficiency at a high cost in hardware design and programming complexity and often lack the flexibility to process different workloads. We propose Squire, a general-purpose accelerator designed to exploit fine-grain parallelism effectively on dependency-bound kernels. Each Squire accelerator has a set of general-purpose low-power in-order cores that can rapidly communicate among themselves and directly access data from the L2 cache. Our proposal integrates one Squire accelerator per core in a typical multicore system, allowing the acceleration of dependency-bound kernels within parallel tasks with minimal software changes. As a case study, we evaluate Squire's effectiveness by accelerating five kernels that implement complex dependency patterns. We use three of these kernels to build an end-to-end read-mapping tool that will be used to evaluate Squire. Squire obtains speedups up to 7.64$ imes$ in dynamic programming kernels. Overall, Squire provides an acceleration for an end-to-end application of 3.66$ imes$. In addition, Squire reduces energy consumption by up to 56% with a minimal area overhead of 10.5% compared to a Neoverse-N1 baseline.

Problem

• Accelerating dependency-bound kernels with complex data patterns
• Overcoming limitations of traditional accelerators in fine-grain parallelism
• Providing efficient acceleration without custom hardware design complexity

Innovation

• Uses low-power cores with rapid inter-core communication
• Integrates accelerators per core for minimal software changes
• Exploits fine-grain parallelism in dependency-bound kernels