High-Performance Computing (HPC) processors are nowadays integrated Cyber-Physical Systems demanding complex and high-bandwidth closed-loop power and thermal control strategies. To efficiently satisfy real-time multi-input multi-output (MIMO) optimal power requirements, high-end processors integrate an on-die power controller system (PCS). While traditional PCSs are based on a simple microcontroller (MCU)-class core, more scalable and flexible PCS architectures are required to support advanced MIMO control algorithms for managing the ever-increasing number of cores, power states, and process, voltage, and temperature variability. This paper presents ControlPULP, an open-source, HW/SW RISC-V parallel PCS platform consisting of a single-core MCU with fast interrupt handling coupled with a scalable multi-core programmable cluster accelerator and a specialized DMA engine for the parallel acceleration of real-time power management policies. ControlPULP relies on FreeRTOS to schedule a reactive power control firmware (PCF) application layer. We demonstrate ControlPULP in a power management use-case targeting a next-generation 72-core HPC processor.We first show that the multi-core cluster accelerates the PCF, achieving 4.9x speedup compared to single-core execution, enabling more advanced power management algorithms within the control hyper-period at a shallow area overhead, about 0.1\% the area of a modern HPC CPU die. We then assess the PCS and PCF by designing an FPGA-based, closed-loop emulation framework that leverages the heterogeneous SoCs paradigm, achieving DVFS tracking with a mean deviation within 3\% the plant's thermal design power (TDP) against a software-equivalent model-in-the-loop approach. Finally, we show that the proposed PCF compares favorably with an industry-grade control algorithm under computational-intensive workloads.

No abstract

No abstract

No abstract

In recent years, new classes of highly dynamic, complex systems are gaining momentum. These systems are characterized by the need to express behaviors driven by external and/or internal changes, i.e. they are reactive and context-aware. These classes include, but are not limited to IoT, smart cities, cyber-physical systems and sensor networks. An important design feature of these systems should be the ability of adapting their behavior to environment changes. This requires handling a runtime representation of the context enriched with variation points that relate different behaviors to possible changes of the representation. In this paper, we present a reference architecture for reactive, context-aware systems able to handle contextual knowledge (that defines what the system perceives) by means of virtual sensors and able to react to environment changes by means of virtual actuators, both represented in a declarative manner through semantic web technologies. To improve the ability to react with a proper behavior to context changes (e.g. faults) that may influence the ability of the system to observe the environment, we allow the definition of logical sensors and actuators through an extension of the SSN ontology (a W3C standard). In our reference architecture a knowledge base of sensors and actuators (hosted by an RDF triple store) is bound to real world by grounding semantic elements to physical devices via REST APIs. The proposed architecture along with the defined ontology try to address the main problems of dynamically reconfigurable systems by exploiting a declarative, queryable approach to enable runtime reconfiguration with the help of (a) semantics to support discovery in heterogeneous environment, (b) composition logic to define alternative behaviors for variation points, (c) bi-causal connection life-cycle to avoid dangling links with the external environment. The proposal is validated in a case study aimed at designing an edge node for smart buildings dedicated to cultural heritage preservation.