Search

The power density in modern Integrated Circuits (ICs) is tremendous. For example, Multi-Processor Systems-on-Chip (MPSoCs) nowadays undergo temperature swings of 40 degrees in 100 milliseconds or less, with rapidly emerging and vanishing sub-millimeter hot spots. As such, not only a simulation-based cooling assessment is vital, but one has to simulate the on-chip thermal phenomena jointly with the heat dissipation system - historically, a challenge. In recent years, however, the idea of coupling traditional 3D chip simulators with heat dissipation models written in Equation-Based Modeling (EBM) languages has proven to be a game changer. EBM languages allow one to compose a model by assembling components described in terms of Differential and Algebraic Equations (DAE) and have the simulation code generated automatically. In this article, we take a tutorial viewpoint on the matter just sketched, to put the reader in the position of exploiting the above technology. We also present the first nucleus of a model library for cooling systems, that we release as free software for the scientific and engineering community. [ABSTRACT FROM AUTHOR]

We propose a periodic event-based control scheme in which, when the sensor triggers an event, also past and subsequent values of the controlled variable are transmitted. We employ this idea to obtain a stability condition independent of the used event-triggering mechanism. In turn, we exploit this independence to devise a methodology for extending existing controller (auto)tuning rules to the event-based setting and provide an example with the internal model control PID (IMC-PID) rule. We then abstract the functionalities that a wireless sensor node has to offer for implementing our scheme and finally present such a node, physically realized in a completely hardware-based form to the advantage of both energy efficiency and security. Simulations results and experiments with the so-obtained control architecture (wireless sensor node plus event-based IMC-PID) are reported. [ABSTRACT FROM AUTHOR]

The density of modern microprocessors is so high, that operating all their units at full power would destroy them by thermal runaway. Hence, thermal control is vital, but at the same time has to integrate with power/performance management, to not unduly limit computational speed. In addition, the controller must be simple and computationally light, as millisecond-scale response is required. Finally, since microprocessors face a variety of operating conditions, postsilicon tuning is an issue. We here present a solution, by exploiting event-based control and a hardware/software partition to maximize efficiency, lightness, and flexibility. We show experiments on real hardware, evidencing the obtained advantages over the state of the art. [ABSTRACT FROM PUBLISHER]

Time synchronization is crucial for wireless sensor networks (WSNs), where operations often rely on time ordering of events. WSNs are deployed in different scenarios, and therefore their timing requirements are often related to the peculiar characteristics of the specific environment they have to act in. Synchronization is anyway always an issue: transactional applications need monotonicity of the nodes’ clocks to avoid time reversal, ultralow power applications call for minimal overhead to allow for low-duty-cycle operation, applications facing extreme environments have to maintain the needed precision in the presence of unforeseen thermal drift, and so on. Specially, control applications on battery-powered devices, where timing is an issue and low-power operation is highly desired, benefit from synchronization. However, to date, the problem of synchronization has been differently faced depending on the application domain. This paper proposes a general solution to the problem of synchronization in WSNs, which seamlessly integrates with the radio stack. In addition, it guarantees monotonic and continuous node clocks with low overhead for the infrastructure. The solution is based on a decentralized control scheme that is stable and robust to thermal stress, without the need for temperature measurements. The control scheme is simulated and implemented on real WSN nodes. The efficiency of the scheme is evaluated with simulations and experiments, providing insights on the maximum synchronization error between nodes, on the communication overhead, and on the limited nodes’ power consumption. The solution is also compared with state-of-the-art alternatives. [ABSTRACT FROM PUBLISHER]