NASCaps

ALOHA: A Unified Platform-Aware Evaluation Method for CNNs Execution on Heterogeneous Systems at the Edge

“…Multiobjective NAS has been an active research topic during the last several years, and a large number of methods, capable of evaluating platform-dependent CNN metrics to assist it, have been proposed. [14] memory, latency, energy Roofline [30] latency low/ medium high X Specialized analytical methods [23] latency, energy, memory medium/ high low X [21], [31] latency, energy [19], [20], [22] latency, energy, throughput medium Measurements [32] latency high very low X [33] latency, energy Look-up tables (LUT) [34] latency high low X [35] latency high medium/ low ML [36] latency, energy high very low X [26] latency ALOHA this work latency, energy, throughput medium/ high high Every evaluation methods category is supplied with a list of methods belonging to it (Column 2), and characterized with: 1) the method accuracy (Column 4); 2) the method re-usability (Column 5); 3) modularity (Column 6). The method re-usability determines how sensitive the evaluation method category is to a specific CNN architecture or/and hardware platform, and determines the applicability of the evaluation method category.…”

“…However, most tools reported in the literature have a limited degree of platform awareness: they fail to capture the effect of potential design choices on the performance metrics achievable by a CNN architecture under consideration executed on a target computing platform, especially when dealing with more complex processing systems, endowed with accelerators, highly-parallel processors and/or GPUs. Estimation methods implemented in these tools are inaccurate ( [13]- [15]), or not sufficiently general ( [16]- [23]), or require a lot of design experiments and modeling skills to be used ( [24]- [26]).…”