NeuroVM: Dynamic Neuromorphic Hardware Virtualization

TL;DR

NeuroVM presents a dynamic virtualization system for neuromorphic hardware, enabling scalable, adaptable, and efficient processing of complex tasks across multiple virtual machines with insights into performance and energy consumption.

Contribution

This work introduces a novel dynamic virtualization architecture for neuromorphic hardware, allowing flexible resource allocation and reconfiguration for diverse applications.

Findings

Scalable throughput up to 5.1 Gib/s across configurations

Energy consumption increases nearly linearly with accelerators

Partial reconfiguration reduces reconfiguration time significantly

Abstract

This paper introduces a novel approach in neuromorphic computing, integrating heterogeneous hardware nodes into a unified, massively parallel architecture. Our system transcends traditional single-node constraints, harnessing the neural structure and functionality of the human brain to efficiently process complex tasks. We present an architecture that dynamically virtualizes neuromorphic resources, enabling adaptable allocation and reconfiguration for various applications. Our evaluation, using diverse applications and performance metrics, provides significant insights into the system's adaptability and efficiency. We observed scalable throughput increases across configurations of 1, 2, and 4 Virtual Machines (VMs), reaching up to 5.1 Gibibits per second (Gib/s) for different data transfer sizes. This scalability demonstrates the system's capacity to handle tasks that require substantial amounts of data. The energy consumption of our virtualized accelerator environment increased nearly linearly with the addition of more NeuroVM accelerators, ranging from 25 to 45 millijoules (mJ) as the number of accelerators increased from 1 to 20. Further, our investigation of reconfiguration overheads revealed that partial reconfigurations significantly reduce the time spent on reconfigurations compared to full reconfigurations, particularly when there are more virtual machines, as indicated by the logarithmic scale of time measurements.