Mitigating Memory-induced Dark Silicon in Many-Accelerator Architectures
Many-Accelerator (MA) systems have been introduced as a promising architectural paradigm that can boost performance and improve power of general-purpose computing platforms. In this paper, the authors focus on the problem of resource under-utilization, i.e. Dark Silicon, in FPGA (Field Programmable Gate Array)-based MA platforms. They show that except the typically expected peak power budget, on-chip memory resources form a severe under-utilization factor in MA platforms, leading up to 75% of dark silicon. Recognizing that static memory allocation - the de-facto mechanism supported by modern design techniques and synthesis tools - forms the main source of memory-induced Dark Silicon; they introduce a novel framework that extends conventional High Level Synthesis (HLS) with Dynamic Memory Management (DMM) features.