Implementing slot-based task-splitting multiprocessor scheduling

Consider the problem of scheduling a set of sporadic tasks on a multiprocessor system to meet deadlines using a task-splitting scheduling algorithm. Task-splitting (also called semi-partitioning) scheduling algorithms assign most tasks to just one processor but a few tasks are assigned to two or more processors, and they are dispatched in a way that ensures that a task never executes on two or more processors simultaneously. A particular type of task-splitting algorithms, called slot-based task-splitting dispatching, is of particular interest because of its ability to schedule tasks with high processor utilizations. Unfortunately, no slot-based task-splitting algorithm has been implemented in a real operating system so far. In this paper we discuss and propose some modifications to the slot-based task-splitting algorithm driven by implementation concerns, and we report the first implementation of this family of algorithms in a real operating system running Linux kernel version 2.6.34. We have also conducted an extensive range of experiments on a 4-core multicore desktop PC running task-sets with utilizations of up to 88%. The results show that the behavior of our implementation is in line with the theoretical framework behind it.

Practical aspects of slot-based task- splitting dispatching in its schedulability analysis

Consider the problem of scheduling a set of sporadic tasks on a multiprocessor system to meet deadlines using a tasksplitting scheduling algorithm. Task-splitting (also called semipartitioning) scheduling algorithms assign most tasks to just one processor but a few tasks are assigned to two or more processors, and they are dispatched in a way that ensures that a task never executes on two or more processors simultaneously. A certain type of task-splitting algorithms, called slot-based task-splitting, is of particular interest because of its ability to schedule tasks at high processor utilizations. We present a new schedulability analysis for slot-based task-splitting scheduling algorithms that takes the overhead into account and also a new task assignment algorithm.

A 12*(1+|R|/(4m))-speed algorithm for scheduling constrained-deadline sporadic real-time tasks on a multiprocessor comprising m processors where a task may request one of |R| sequentially-reusable shared resources

We present a 12*(1+|R|/(4m))-speed algorithm for scheduling constrained-deadline sporadic real-time tasks on a multiprocessor comprising m processors where a task may request one of |R| sequentially-reusable shared resources.

Implementing multicore real-time scheduling algorithms based on task splitting using ada 2012

Multiprocessors, particularly in the form of multicores, are becoming standard building blocks for executing reliable software. But their use for applications with hard real-time requirements is non-trivial. Well-known realtime scheduling algorithms in the uniprocessor context (Rate-Monotonic [1] or Earliest-Deadline-First [1]) do not perform well on multiprocessors. For this reason the scientific community in the area of real-time systems has produced new algorithms specifically for multiprocessors. In the meanwhile, a proposal [2] exists for extending the Ada language with new basic constructs which can be used for implementing new algorithms for real-time scheduling; the family of task splitting algorithms is one of them which was emphasized in the proposal [2]. Consequently, assessing whether existing task splitting multiprocessor scheduling algorithms can be implemented with these constructs is paramount. In this paper we present a list of state-of-art task-splitting multiprocessor scheduling algorithms and, for each of them, we present detailed Ada code that uses the new constructs.

Optimising and adapting the QoS of a dynamic set of inter-dependent tasks

Due to the growing complexity and adaptability requirements of real-time systems, which often exhibit unrestricted Quality of Service (QoS) inter-dependencies among supported services and user-imposed quality constraints, it is increasingly difficult to optimise the level of service of a dynamic task set within an useful and bounded time. This is even more difficult when intending to benefit from the full potential of an open distributed cooperating environment, where service characteristics are not known beforehand and tasks may be inter-dependent. This paper focuses on optimising a dynamic local set of inter-dependent tasks that can be executed at varying levels of QoS to achieve an efficient resource usage that is constantly adapted to the specific constraints of devices and users, nature of executing tasks and dynamically changing system conditions. Extensive simulations demonstrate that the proposed anytime algorithms are able to quickly find a good initial solution and effectively optimise the rate at which the quality of the current solution improves as the algorithms are given more time to run, with a minimum overhead when compared against their traditional versions.

Handling shared resources and precedence constraints in open systems

There is an increasing demand for highly dynamic realtime systems where several independently developed applications with different timing requirements can coexist. This paper proposes a protocol to integrate shared resources and precedence constraints among tasks in such systems assuming no precise information on critical sections and computation times is available. The concept of bandwidth inheritance is combined with a capacity sharing and stealing mechanism to efficiently exchange bandwidth among needed tasks, minimising the cost of blocking.

Scheduling arbitrary-deadline sporadic task systems on multiprocessors

A new algorithm is proposed for scheduling preemptible arbitrary-deadline sporadic task systems upon multiprocessor platforms, with interprocessor migration permitted. This algorithm is based on a task-splitting approach - while most tasks are entirely assigned to specific processors, a few tasks (fewer than the number of processors) may be split across two processors. This algorithm can be used for two distinct purposes: for actually scheduling specific sporadic task systems, and for feasibility analysis. Simulation- based evaluation indicates that this algorithm offers a significant improvement on the ability to schedule arbitrary- deadline sporadic task systems as compared to the contemporary state-of-art. With regard to feasibility analysis, the new algorithm is proved to offer superior performance guarantees in comparison to prior feasibility tests.

Dynamic QoS adaptation of inter- dependent task sets in cooperative embedded systems

Due to the growing complexity and dynamism of many embedded application domains (including consumer electronics, robotics, automotive and telecommunications), it is increasingly difficult to react to load variations and adapt the system's performance in a controlled fashion within an useful and bounded time. This is particularly noticeable when intending to benefit from the full potential of an open distributed cooperating environment, where service characteristics are not known beforehand and tasks may exhibit unrestricted QoS inter-dependencies. This paper proposes a novel anytime adaptive QoS control policy in which the online search for the best set of QoS levels is combined with each user's personal preferences on their services' adaptation behaviour. Extensive simulations demonstrate that the proposed anytime algorithms are able to quickly find a good initial solution and effectively optimise the rate at which the quality of the current solution improves as the algorithms are given more time to run, with a minimum overhead when compared against their traditional versions.

Handling QoS dependencies in distributed cooperative real-time systems

Due to the growing complexity and adaptability requirements of real-time embedded systems, which often exhibit unrestricted inter-dependencies among supported services and user-imposed quality constraints, it is increasingly difficult to optimise the level of service of a dynamic task set within an useful and bounded time. This is even more difficult when intending to benefit from the full potential of an open distributed cooperating environment, where service characteristics are not known beforehand. This paper proposes an iterative refinement approach for a service’s QoS configuration taking into account services’ inter-dependencies and quality constraints, and trading off the achieved solution’s quality for the cost of computation. Extensive simulations demonstrate that the proposed anytime algorithm is able to quickly find a good initial solution and effectively optimises the rate at which the quality of the current solution improves as the algorithm is given more time to run. The added benefits of the proposed approach clearly surpass its reducedoverhead.

Shared resources and precedence constraints with capacity sharing and stealing

This paper proposes a new strategy to integrate shared resources and precedence constraints among real-time tasks, assuming no precise information on critical sections and computation times is available. The concept of bandwidth inheritance is combined with a greedy capacity sharing and stealing policy to efficiently exchange bandwidth among tasks, minimising the degree of deviation from the ideal system's behaviour caused by inter-application blocking. The proposed capacity exchange protocol (CXP) focus on exchanging extra capacities as early, and not necessarily as fairly, as possible. This loss of optimality is worth the reduced complexity as the protocol's behaviour nevertheless tends to be fair in the long run and outperforms other solutions in highly dynamic scenarios, as demonstrated by extensive simulations.

Competitive dynamics for behavior coordination in a joint transportation task

We address the problem of coordinating two non-holonomic mobile robots that move in formation while transporting a long payload. A competitive dynamics is introduced that gradually controls the activation and deactivation of individual behaviors. This process introduces (asymmetrical) hysteresis during behavioral switching. As a result behavioral oscillations, due to noisy information, are eliminated. Results in indoor environments show that if parameter values are chosen within reasonable ranges then, in spite of noise in the robots communi- cation and sensors, the overall robotic system works quite well even in cluttered environments. The robots overt behavior is stable and smooth.

Dynamic adaptation of stability periods for service level agreements

A QoS adaptation to dynamically changing system conditions that takes into consideration the user’s constraints on the stability of service provisioning is presented. The goal is to allow the system to make QoS adaptation decisions in response to fluctuations in task traffic flow, under the control of the user. We pay special attention to the case where monitoring the stability period and resource load variation of Service Level Agreements for different types of services is used to dynamically adapt future stability periods, according to a feedback control scheme. System’s adaptation behaviour can be configured according to a desired confidence level on future resource usage. The viability of the proposed approach is validated by preliminary experiments.

Improvement of a filters method in a derivative free optimization

Constraints nonlinear optimization problems can be solved using penalty or barrier functions. This strategy, based on solving the problems without constraints obtained from the original problem, have shown to be e ective, particularly when used with direct search methods. An alternative to solve the previous problems is the lters method. The lters method introduced by Fletcher and Ley er in 2002, , has been widely used to solve problems of the type mentioned above. These methods use a strategy di erent from the barrier or penalty functions. The previous functions de ne a new one that combine the objective function and the constraints, while the lters method treat optimization problems as a bi-objective problems that minimize the objective function and a function that aggregates the constraints. Motivated by the work of Audet and Dennis in 2004, using lters method with derivative-free algorithms, the authors developed works where other direct search meth- ods were used, combining their potential with the lters method. More recently. In a new variant of these methods was presented, where it some alternative aggregation restrictions for the construction of lters were proposed. This paper presents a variant of the lters method, more robust than the previous ones, that has been implemented with a safeguard procedure where values of the function and constraints are interlinked and not treated completely independently.

A hybrid genetic algorithm for the multi-mode resource-constrained project scheduling

A construction project is a group of discernible tasks or activities that are conduct-ed in a coordinated effort to accomplish one or more objectives. Construction projects re-quire varying levels of cost, time and other resources. To plan and schedule a construction project, activities must be defined sufficiently. The level of detail determines the number of activities contained within the project plan and schedule. So, finding feasible schedules which efficiently use scarce resources is a challenging task within project management. In this context, the well-known Resource Constrained Project Scheduling Problem (RCPSP) has been studied during the last decades. In the RCPSP the activities of a project have to be scheduled such that the makespan of the project is minimized. So, the technological precedence constraints have to be observed as well as limitations of the renewable resources required to accomplish the activities. Once started, an activity may not be interrupted. This problem has been extended to a more realistic model, the multi-mode resource con-strained project scheduling problem (MRCPSP), where each activity can be performed in one out of several modes. Each mode of an activity represents an alternative way of combining different levels of resource requirements with a related duration. Each renewable resource has a limited availability for the entire project such as manpower and machines. This paper presents a hybrid genetic algorithm for the multi-mode resource-constrained pro-ject scheduling problem, in which multiple execution modes are available for each of the ac-tivities of the project. The objective function is the minimization of the construction project completion time. To solve the problem, is applied a two-level genetic algorithm, which makes use of two separate levels and extend the parameterized schedule generation scheme. It is evaluated the quality of the schedules and presents detailed comparative computational re-sults for the MRCPSP, which reveal that this approach is a competitive algorithm.

Project scheduling under multiple resources constraints using a genetic algorithm

The resource constrained project scheduling problem (RCPSP) is a difficult problem in combinatorial optimization for which extensive investigation has been devoted to the development of efficient algorithms. During the last couple of years many heuristic procedures have been developed for this problem, but still these procedures often fail in finding near-optimal solutions. This paper proposes a genetic algorithm for the resource constrained project scheduling problem. The chromosome representation of the problem is based on random keys. The schedule is constructed using a heuristic priority rule in which the priorities and delay times of the activities are defined by the genetic algorithm. The approach was tested on a set of standard problems taken from the literature and compared with other approaches. The computational results validate the effectiveness of the proposed algorithm.