Allocating one common accelerator-rich platform for many streaming applications

Abstract

Many demanding streaming applications share functional and structural similarities with other applications in their respective domain, e.g. video analytics, software-defined radio, and radar. This opens the opportunity for specialization (e.g. heterogeneous computing) to achieve the needed efficiency and/or performance.However, current Design Space Exploration (DSE) focuses on an individual application in isolation (e.g. one particular vision flow), but not a set of similar applications. Hence, optimizations that occur due to considering multiple applications simultaneously are missed. New DSE methodologies and tools are needed with a broader scope of application sets instead of individual applications. This thesis introduces a novel Domain DSE approach focusing on streaming applications. Key contributions are: (1) a formalized method to extract the functional and structural similarities of domain applications, (2) domain application generation to provide enough synthetic domains as study cases, (3) a rapid platform performance estimation and comparison at two abstraction levels: Domain Score (DS) and Analytic Performance Estimation (APE) model, (4) a methodology to evaluate a platform's benefit for a set of applications, and (5) two novel algorithms, Dynamic Score Selection (DSS) and GenetIc Domain Exploration (GIDE), for hardware/software partitioning of a domain-specific platform to maximize the throughput across domain applications (under certain constraints). We demonstrate DSS's and \ga's benefits using OpenVX applications and synthetic domains. The DSS and GIDE generated domain-specific platforms improve performance over application-specific platforms by 58%, and 75% for OpenVX, as well as by 23% and 48% for synthetic applications. GIDE's platforms reach 99.8% (OpenVX) and 97.6% (synthetic) throughput of the domain optimal platform obtained through exhaustive search.