Thidapat Chantem

Temperature-aware scheduling and assignment for hard real-time applications on MPSoCs

Increasing integrated circuit (IC) power densities and temperatures may hamper multiprocessor system-on-chip (MPSoC) use in hard real-time systems. This paper formalizes the temperature-aware real-time MPSoC assignment and scheduling problem and presents an optimal phased steady-state mixed integer linear programming-based solution that considers the impact of scheduling and assignment decisions on MPSoC thermal profiles to directly minimize the chip peak temperature. We also introduce a flexible heuristic framework for task assignment and scheduling that permits system designers to trade off accuracy for running time when solving large problem instances. Finally, for task sets with sufficient slack, we show that inserting idle times between task executions can further reduce the peak temperature of the MPSoC quite significantly.

Temperature-Aware Scheduling and Assignment for Hard Real-Time Applications on MPSoCs

Increasing integrated circuit (IC) power densities and temperatures may hamper multiprocessor system-on-chip (MPSoC) use in hard real-time systems. This paper formalizes the temperature-aware real-time MPSoC assignment and scheduling problem and presents an optimal phased steady-state mixed integer linear programming-based solution that considers the impact of scheduling and assignment decisions on MPSoC thermal profiles to directly minimize the chip peak temperature. We also introduce a flexible heuristic framework for task assignment and scheduling that permits system designers to trade off accuracy for running time when solving large problem instances. Finally, for task sets with sufficient slack, we show that inserting idle times between task executions can further reduce the peak temperature of the MPSoC quite significantly.

On self-triggered full-information H-infinity controllers

A self-triggered control task is one in which the task determines its next release time. It has been conjectured that self-triggering can relax the requirements on a real-time scheduler while maintaining application (i.e. control system) performance. This paper presents preliminary results supporting that conjecture for a self-triggered real-time system implementing full-information H∞ controllers. Release times are selected to enforce upper bounds on the induced L2 gain of a linear feedback control system. These release times are treated as requests by the system scheduler, which then assigns actual release times using Buttazzos elastic scheduling algorithm. Preliminary experimental results from a Matlab stateflow simulink model demonstrated a remarkable robustness to scheduling delays induced by real-time schedulers. These results show that self-triggered controllers are indeed able to maintain acceptable levels of application performance during prolonged periods of processor overloading.

Online work maximization under a peak temperature constraint

Increasing power densities and the high cost of low thermal resistance packages and cooling solutions make it impractical to design processors for worst-case temperature scenarios. As a result, packages and cooling solutions are designed for less than worst-case power densities and dynamic voltage and frequency scaling (DVFS) is used to prevent dangerous on-chip temperatures at run time. Unfortunately, DVFS can cause unpredicted drops in performance (e.g., long response times). We propose and optimally solve the problem of thermally-constrained online work maximization for general-purpose computing systems on uniprocessors with discrete speed levels and non-negligible transition overheads. Simulation results show that our approach completes 47.7% on average and up to 68.0% more cycles than a naive policy.

System-level reliability modeling for MPSoCs

The reliability of multi-processor systems-on-chip (MPSoCs) is affected by several inter-dependent system-level and physical effects. Accurate and fast reliability modeling is a primary challenge in the design and optimization of reliable MPSoCs. This paper presents a reliability modeling framework that integrates device-, component-, and system-level models. This framework contains modules for electromigration, time-dependent dielectric breakdown, stress migration, and variable-amplitude thermal cycling. A new statistical reliability distribution is proposed for accurate characterization of components containing too few devices for an extreme value distribution to be appropriate. A hierarchical system-level survival lattice based Monte Carlo technique is used to estimate the temporal fault distributions of MPSoCs that use arbitrary static and dynamic reliability-enhancing redundancy schemes. Physical process variation, which may have a significant impact on MPSoC reliability, is considered in the model. The proposed modeling technique has 5% average error in mean time to failure and reduces simulation time by nearly 3 orders of magnitude relative to a non-hierarchical Monte Carlo technique.

Generalized Elastic Scheduling for Real-Time Tasks

The elastic task model is a powerful model for adapting periodic real-time systems in the presence of uncertainty. This work generalizes the existing elastic scheduling approach in several directions. First, it presents a general framework, which formulates a trade-off between task schedulability and a specific performance metric as an optimization problem. Such a framework allows real-time systems under overloads to graciously adapt by adjusting their performance level. Second, it is shown in this work that the well-known task compression algorithm in fact solves a quadratic programming problem that seeks to minimize the sum of the squared deviation of a tasks utilization from initial desired utilization. This finding indicates that the task compression algorithm may be applied to efficiently solve other similar types of problems that often arise in real-time applications. In particular, an iterative approach is proposed to solve the period selection problem for real-time tasks with deadlines less than respective periods. Further, the framework is adapted to solve the deadline selection problem, which is useful in some control systems with fixed periods.

Enhancing multicore reliability through wear compensation in online assignment and scheduling

System reliability is a crucial concern especially in multicore systems which tend to have high power density and hence temperature. Existing reliability-aware methods are either slow and non-adaptive (offline techniques) or do not use task assignment and scheduling to compensate for uneven core wear states (online techniques). In this article, we present a dynamically-activated task assignment and scheduling algorithm based on theoretical results that explicitly optimizes system life-time. We also propose a data distillation method that dramatically reduces the size of the thermal profiles to make full system reliability analysis viable online. Simulation results show that our algorithm results in between 27–291% improvement to system lifetime compared to existing techniques for four-core systems.

Generalized Elastic Scheduling

The elastic task model (Buttazzo et al., 2002) is a powerful model for adapting real-time systems in the presence of uncertainty. This paper generalizes the existing elastic scheduling approach in several directions. It reveals that the original task compression algorithm in (Buttazzo et al., 2002) in fact solves a quadratic programming problem that seeks to minimize the sum of the squared deviation of a tasks utilization from initial desired utilization. This finding indicates that the task compression algorithm may be applied to efficiently solve other similar types of problems. In particular, an iterative approach is proposed to solve the task compression problem for real-time tasks with deadlines less than respective periods. Furthermore, a new objective for minimizing the average difference of task periods from desired values is introduced and a closed-form formula is derived for solving the problem without recursion

Meeting End-to-End Deadlines through Distributed Local Deadline Assignments

In a distributed real-time system, jobs are often executed on a number of processors and must be completed by their end-to-end deadlines. Without considering resource competition among different jobs on each processor, deadline requirements may be violated. The paper introduces a distributed approach to assigning local deadlines to the jobs on each processor. The approach leads to improved schedulability results by considering disparate workloads among the processors due to competing jobs having different paths. Simulation results based on randomly generated workloads indicate that the proposed approach outperforms existing work in terms of both the number of feasible task sets (between 22% and 75%) and the number of feasible jobs (between 57% and 46%).

Network-Aware Dynamic Voltage and Frequency Scaling

Reducing energy consumption is an important consideration in embedded real-time system development. This work examines systems that contain a DVFS managed CPU executing packet producing tasks and a DPM-controlled network interface. We introduce a novel approach to minimize energy consumed by the network resource on such a system, through careful selection of voltage and frequency levels on the CPU. Contrary to existing claims which state that DVFS should not be employed when the CPU is not a significant consumer of energy, we show that our DVFS technique can reduce system energy by as much as 35%, even when the CPU energy consumption is negligible. Furthermore, we motivate the need to balance the CPU and network energy and present two techniques to do so. One is based on off-line analysis and the other is a conservative on-line approach. We then validate the proposed methods using both simulation and an implementation in the Linux kernel