The maturity level of RISC-V and the availability of domain-specific instruction set extensions, like vector processing, make RISC-V a good candidate for supporting the integration of specialized hardware in processor cores for the High Performance Computing (HPC) application domain. In this article, 1 we present Vitruvius+, the vector processing acceleration engine that represents the core of vector instruction execution in the HPC challenge that comes within the EuroHPC initiative. It implements the RISC-V vector extension (RVV) 0.7.1 and can be easily connected to a scalar core using the Open Vector Interface standard. Vitruvius+ natively supports long vectors: 256 double precision floating-point elements in a single vector register. It is composed of a set of identical vector pipelines (lanes), each containing a slice of the Vector Register File and functional units (one integer, one floating point). The vector instruction execution scheme is hybrid in-order/out-of-order and is supported by register renaming and arithmetic/memory instruction decoupling. On a stand-alone synthesis, Vitruvius+ reaches a maximum frequency of 1.4 GHz in typical conditions (TT/0.80V/25°C) using GlobalFoundries 22FDX FD-SOI. The silicon implementation has a total area of 1.3 mm 2 and maximum estimated power of ∼920 mW for one instance of Vitruvius+ equipped with eight vector lanes.
22 April 2025 in IEEE Computer Architecture Letters
Graph Pattern Mining (GPM) algorithms extract meaningful information within graph structures, making them fundamental building blocks for multiple application domains. However, their performance is bottlenecked by hard-to-predict divergence control, cache pollution, and low parallelism caused by index matching operations that dominate the execution time. To address these challenges, this paper introduces Halis, a hardware-software co-designed Near-Cache Accelerator for GPM workloads on commercial multi-core CPUs. By executing index matching operations near the Last-Level Cache (LLC), Halis reduces data movement and cache pollution in upper cache levels while minimizing divergence control and enhancing parallelism. To achieve this, Halis repurposes underutilized Content Addressable Memories (CAMs) in hardware data prefetchers, taking advantage of their efficient lookup capabilities for GPM workloads. Furthermore, Halis includes virtual memory support, ensuring compatibility with commodity operating systems. Designed as a decoupled programmable accelerator, it operates via memory-mapped registers. Our evaluation demonstrates that Halis outperforms software and hardware approaches by 26.9× and 2.4× respectively, while incurring a negligible area overhead of 0.05% over the CPU baseline.