Alfons Crespo

Mode Change Protocols for Real-Time Systems: A Survey and a New Proposal

This paper contains both a survey of mode change protocols for single-processor, fixed-priority, preemptively scheduled real-time systems, and a proposal of several new protocols along with their corresponding schedulability analysis and configuration methods. First, a classification of the protocols found in the literature is given and a set of requirements is proposed for their evaluation. Then, the new protocols are introduced and discussed in the light of the stated requirements. A number of mode change protocols are based on delaying the initiation of the new mode by applying an offset to the initial release of tasks in the destination mode. We tackle the problem of how to obtain these offsets for the proposed protocols. The issue of consistently sharing resources during the mode change by means of a priority inheritance protocol is also dealt with.

Improvement in feasibility testing for real-time tasks

Scheduling analysis in real-time systems require an off-line feasibility (schedulability) test to guarantee the response time of critical tasks. There are fast and efficient tests for static schedulers. However, when a dynamic scheduler is required, the available tests are not as feast and efficient as static ones. In this paper, two different characterizations of feasible task sets are presented. These characterizations lead to an new feasibility algorithm. The proposed algorithm has an worst-case exponential complexity, but experimental results indicated that it runs on pseudo-polynomial time for a very large percentage of task sets. The algorithm also provides a sufficient condition for feasible asynchronous task sets. One of the main contributions of this work is the theoretical approach used to obtain the new feasibility test. The results of a large number of simulations, as well as, the theoretical demonstrations point out the improvements reached over previous tests.

TLSF: a new dynamic memory allocator for real-time systems

Dynamic storage allocation (DSA) algorithms play an important role in the modern software engineering paradigms and techniques (such as object oriented programming). Using DSA increases the flexibility and functionalities of applications. There exists a large number of references to this particular issue in the literature. However, the use of DSA has been considered a source of indeterminism in the real-time domain, due to the unconstrained response time of DSA algorithms and the fragmentation problem. Nowadays, real-time applications require more flexibility: the ability to adjust system configuration in response to workload changes and application reconfiguration. This aspect adds value to the definition and implementation of dynamic storage allocation algorithms. Considering these reasons, DSA algorithms with a bounded and acceptable timing behaviour must be developed to be used by real-time operating systems (RTOSs). In this paper a DSA algorithm called two-level segregated fit memory allocator (TLSF), developed specifically to be used by RTOS, is introduced. The TLSF algorithm provides explicit allocation and deallocation of memory blocks with a temporal cost /spl Theta/(1).

IRIS: a new reclaiming algorithm for server-based real-time systems

In this paper we present a new algorithm for CPU resource reservation in real-time systems that allows the coexistence of hard, soft and non real-time tasks. The proposed algorithm is specifically designed to handle computational overload. A task that needs more CPU-time than reserved can reuse the spare bandwidth, without interfering with the others tasks. With respect to other reclamation schemes, the novelty of the proposed algorithm is that the spare bandwidth is fairly distributed among the needing servers. The effectiveness of the algorithm is demonstrated with an extensive set of experiments. We also propose a methodology to set scheduling parameters depending on the type of the task and on the time constraints needed.

An optimal algorithm for scheduling soft aperiodic tasks in dynamic-priority preemptive systems

The paper addresses the problem of jointly scheduling tasks with both hard and soft real time constraints. We present a new analysis applicable to systems scheduled using a priority preemptive dispatcher, with priorities assigned dynamically according to the EDF policy. Further, we present a new efficient online algorithm (the acceptor algorithm) for servicing aperiodic work load. The acceptor transforms a soft aperiodic task into a hard one by assigning a deadline. Once transformed, aperiodic tasks are handled in exactly the same way as periodic tasks with hard deadlines. The proposed algorithm is shown to be optimal in terms of providing the shortest aperiodic response time among fixed and dynamic priority schedulers. It always guarantees the proper execution of periodic hard tasks. The approach is composed of two parts: an offline analysis and a run time scheduler. The offline algorithm runs in pseudopolynomial time O(mn), where n is the number of hard periodic tasks and m is the hyperperiod/min deadline.

Real-time control of non-uniformly sampled systems

Abstract Industrial applications of digital control require a synergy between well-designed control algorithms and carefully implemented control systems. The design of digital controllers should not be constrained to the simplest case of a single rate with sychronous sampling strategies, and the digital implementation should take account of the actual real-time operating environment. In this paper, control design techniques under non-uniform sampling schemes are reviewed, and computer improvements to cancel or reduce the effect of delays and process interactions are presented.

RT control scheduling to reduce control performance degrading

In the framework of real-time digital control, two fundamental parameters are defined, the control effort and the control action interval. The first one is related to the strength of the control that, due to the intersampling open-loop control, determines the degrading of performances under unexpected delays. The second one refers to the unavoidable delays in the multitasking environment due to interactions among the tasks. As a consequence, the scheduling policy should consider not only the tasks delays but also their influence in the control loop behavior, being calculated to minimize the overall degrading of performances.

Reducing Delays in RT Control: The Control Action Interval

Abstract Industrial application of digital control requires the synergy between well designed control algorithms and carefully implemented control systems. The design of digital controllers considers different parameters when selecting the appropriate algorithms. However, the control performances can be reduced if the system is implemented with several control loops (tasks) which are not considered in the design phase. In this paper, we analyze the control theory and the effects of the multitasking control loops from the data acquisition and output actuation. The paper shows how the output actuation can vary significantly and proposes a method to reduce the jitter effects and the parameters to schedule the tasks. This reduced action interval (Control Action Interval) can be considered in the control design phase in order to properly adjust the control algorithm.

Using exact feasibility tests for allocating real-time tasks in multiprocessor systems

The paper introduces improvements in partitioning schemes for multiprocessor real time systems which allow higher processor utilization and enhanced schedulability by using exact feasibility tests to evaluate the schedulability limit of a processor. The paper analyzes how to combine these tests with existing bin packing algorithms for processor allocation and provides new variants which are exhaustively evaluated using two assumptions: variable and fixed number of processors. The problem of evaluating these algorithms is complex, since metrics are usually based on comparing the performance of a given algorithm to an optimal one, but determining optimals is often NP hard on multiprocessors. This problem has been overcome by defining lower or upper bounds on the performance of an optimal algorithm and then defining metrics with respect these bounds. The evaluation has shown that the algorithms exhibit extremely good behavior and they can be considered very close to the maximum achievable utilization. It is also shown that dynamic priority policies produce significantly better results than fixed priority policies when task sets require high processor utilizations.

Optimal deadline assignment for periodic real-time tasks in dynamic priority systems

Real-time systems are often designed using a set of periodic tasks. Task periods are usually set by the system requirements, but deadlines and computation times can be modified in order to improve system performance. Sensitivity analysis in real-time systems has focused on changes in task computation times using fixed priority analysis. Only a few studies deal with the modification of deadlines in dynamic priority scheduling. The aim of this work is to provide a sensitivity analysis for task deadlines in the context of dynamic-priority, pre-emptive, uniprocessor scheduling. In this paper, we present a deadline minimisation method that achieves the maximum reduction. As undertaken in other studies concerning computation times, we also define and calculate the critical scaling factor for task deadlines. Our proposal is evaluated and compared with other works in terms of jitter. The deadline minimisation can be used to strongly reduce jitter of control tasks, in a real-time control application