Adaptivity Support for MPSoCs Based on Process Migration in Polyhedral Process Networks

Abstract

System adaptivity is becoming an important feature of modern embedded multiprocessor systems. To achieve the goal of system adaptivity when executing Polyhedral Process Networks (PPNs) on a generic tiled Network-on-Chip (NoC) MPSoC platform, we propose an approach to enable the run-time migration of processes among the available platform resources. In our approach, process migration is allowed by a middleware layer which comprises two main components. The first component concerns the inter-tile data communication between processes. We develop and evaluate a number of different communication approaches which implement the semantics of the PPN model of computation on a generic NoC platform. The presented communication approaches do not depend on the mapping of processes and have been implemented on a Network-on-Chip multiprocessor platform prototyped on an FPGA. Their comparison in terms of the introduced overhead is presented in two case studies with different communication characteristics. The second middleware component allows the actual run-time migration of PPN processes. To this end, we propose and evaluate a process migration mechanism which leverages the PPN model of computation to guarantee a predictable and efficient migration procedure. The efficiency and applicability of the proposed migration mechanism is shown in a real-life case study.

Towards Self-Adaptive KPN Applications on NoC-Based MPSoCs

“…The platform is generated by the SHMPI builder tool [15] and it is a mesh-based 2 × 2 NoC consisting of Microblaze processors emulated on a Xilinx Virtex6 FPGA. The software stack enabling the execution of KPN applications on this platform is based on the request-driven middleware explained in [12].…”

“…(2) Our controller is generic and requires minimal knowledge of application characteristics as compared to these methods which require deep knowledge of the data-flow structure of the application. A middleware-based approach to enable run-time migration of processes among tiles of a NoC is presented in [12]. Such a technique helps in achieving application-independent adaptivity support such as fault tolerance.…”

System Adaptivity and Fault-Tolerance in NoC-based MPSoCs: The MADNESS Project Approach

“…Several methods to implement the PPN communication over NoC-based MPSoCs are described in [10], namely Virtual Connector, Virtual Connector with Variable Rate, and Requestdriven. However, in MADNESS we adopt the Request-driven communication approach as it leads to an easier implementation of the migration mechanism due to the reduced number of synchronization points between processes.…”

“…Interrupt-based request messages: In [10] we implemented the Request-driven approach using a passive middleware. This means that the synchronization protocol was implemented by polling the Network Adapter buffer on each tile to fetch incoming requests and then react consequently.…”

A hierarchical run-time adaptive resource allocation framework for large-scale MPSoC systems

“…There are a number of technology enablers toward adaptive systems. For example, there has been significant research attention on techniques for dynamically re-mapping application tasks to different processing resources [2,3,16]. But they are rarely considered for large-scale heterogeneous MPSoCs.…”