Combining RTSJ with Fork/Join: a priority-based model

This paper discusses the increased need to support dynamic task-level parallelism in embedded real-time systems and proposes a Java framework that combines the Real-Time Specification for Java (RTSJ) with the Fork/Join (FJ) model, following a fixed priority-based scheduling scheme. Our work intends to support parallel runtimes that will coexist with a wide range of other complex independently developed applications, without any previous knowledge about their real execution requirements, number of parallel sub-tasks, and when those sub-tasks will be generated.

Competitive analysis of static-priority partitioned scheduling on uniform multiprocessors

Consider the problem of scheduling a set of sporadically arriving tasks on a uniform multiprocessor with the goal of meeting deadlines. A processor p has the speed Sp. Tasks can be preempted but they cannot migrate between processors. On each processor, tasks are scheduled according to rate-monotonic. We propose an algorithm that can schedule all task sets that any other possible algorithm can schedule assuming that our algorithm is given processors that are √2 / √2−1 ≈ 3.41 times faster. No such guarantees are previously known for partitioned static-priority scheduling on uniform multiprocessors.

Global static-priority preemptive multiprocessor scheduling with utilization bound 38%

Consider the problem of scheduling real-time tasks on a multiprocessor with the goal of meeting deadlines. Tasks arrive sporadically and have implicit deadlines, that is, the deadline of a task is equal to its minimum inter-arrival time. Consider this problem to be solved with global static-priority scheduling. We present a priority-assignment scheme with the property that if at most 38% of the processing capacity is requested then all deadlines are met.

The utilization bound of static-priority preemptive partitioned multiprocessor scheduling is 50%

This paper studies static-priority preemptive scheduling on a multiprocessor using partitioned scheduling. We propose a new scheduling algorithm and prove that if the proposed algorithm is used and if less than 50% of the capacity is requested then all deadlines are met. It is known that for every static-priority multiprocessor scheduling algorithm, there is a task set that misses a deadline although the requested capacity is arbitrary close to 50%.

Real-time Task-Splitting scheduling algorithms framework

Nos dias de hoje, os sistemas de tempo real crescem em importância e complexidade. Mediante a passagem do ambiente uniprocessador para multiprocessador, o trabalho realizado no primeiro não é completamente aplicável no segundo, dado que o nível de complexidade difere, principalmente devido à existência de múltiplos processadores no sistema. Cedo percebeu-se, que a complexidade do problema não cresce linearmente com a adição destes. Na verdade, esta complexidade apresenta-se como uma barreira ao avanço científico nesta área que, para já, se mantém desconhecida, e isto testemunha-se, essencialmente no caso de escalonamento de tarefas. A passagem para este novo ambiente, quer se trate de sistemas de tempo real ou não, promete gerar a oportunidade de realizar trabalho que no primeiro caso nunca seria possível, criando assim, novas garantias de desempenho, menos gastos monetários e menores consumos de energia. Este último fator, apresentou-se desde cedo, como, talvez, a maior barreira de desenvolvimento de novos processadores na área uniprocessador, dado que, à medida que novos eram lançados para o mercado, ao mesmo tempo que ofereciam maior performance, foram levando ao conhecimento de um limite de geração de calor que obrigou ao surgimento da área multiprocessador. No futuro, espera-se que o número de processadores num determinado chip venha a aumentar, e como é óbvio, novas técnicas de exploração das suas inerentes vantagens têm de ser desenvolvidas, e a área relacionada com os algoritmos de escalonamento não é exceção. Ao longo dos anos, diferentes categorias de algoritmos multiprocessador para dar resposta a este problema têm vindo a ser desenvolvidos, destacando-se principalmente estes: globais, particionados e semi-particionados. A perspectiva global, supõe a existência de uma fila global que é acessível por todos os processadores disponíveis. Este fato torna disponível a migração de tarefas, isto é, é possível parar a execução de uma tarefa e resumir a sua execução num processador distinto. Num dado instante, num grupo de tarefas, m, as tarefas de maior prioridade são selecionadas para execução. Este tipo promete limites de utilização altos, a custo elevado de preempções/migrações de tarefas. Em contraste, os algoritmos particionados, colocam as tarefas em partições, e estas, são atribuídas a um dos processadores disponíveis, isto é, para cada processador, é atribuída uma partição. Por essa razão, a migração de tarefas não é possível, acabando por fazer com que o limite de utilização não seja tão alto quando comparado com o caso anterior, mas o número de preempções de tarefas decresce significativamente. O esquema semi-particionado, é uma resposta de caráter hibrido entre os casos anteriores, pois existem tarefas que são particionadas, para serem executadas exclusivamente por um grupo de processadores, e outras que são atribuídas a apenas um processador. Com isto, resulta uma solução que é capaz de distribuir o trabalho a ser realizado de uma forma mais eficiente e balanceada. Infelizmente, para todos estes casos, existe uma discrepância entre a teoria e a prática, pois acaba-se por se assumir conceitos que não são aplicáveis na vida real. Para dar resposta a este problema, é necessário implementar estes algoritmos de escalonamento em sistemas operativos reais e averiguar a sua aplicabilidade, para caso isso não aconteça, as alterações necessárias sejam feitas, quer a nível teórico quer a nível prá

Adapting mode switches into the hierarchical scheduling

Mode switches are used to partition the system’s behavior into different modes to reduce the complexity of large embedded systems. Such systems operate in multiple modes in which each one corresponds to a specific application scenario; these are called Multi-Mode Systems (MMS). A different piece of software is normally executed for each mode. At any given time, the system can be in one of the predefined modes and then be switched to another as a result of a certain condition. A mode switch mechanism (or mode change protocol) is used to shift the system from one mode to another at run-time. In this thesis we have used a hierarchical scheduling framework to implement a multi-mode system called Multi-Mode Hierarchical Scheduling Framework (MMHSF). A two-level Hierarchical Scheduling Framework (HSF) has already been implemented in an open source real-time operating system, FreeRTOS, to support temporal isolation among real-time components. The main contribution of this thesis is the extension of the HSF featuring a multimode feature with an emphasis on making minimal changes in the underlying operating system (FreeRTOS) and its HSF implementation. Our implementation uses fixed-priority preemptive scheduling at both local and global scheduling levels and idling periodic servers. It also now supports different modes of the system which can be switched at run-time. Each subsystem and task exhibit different timing attributes according to mode, and upon a Mode Change Request (MCR) the task-set and timing interfaces of the entire system (including subsystems and tasks) undergo a change. A Mode Change Protocol specifies precisely how the system-mode will be changed. However, an application may not only need to change a mode but also a different mode change protocol semantic. For example, the mode change from normal to shutdown can allow all the tasks to be completed before the mode itself is changed, while changing a mode from normal to emergency may require aborting all tasks instantly. In our work, both the system mode and the mode change protocol can be changed at run-time. We have implemented three different mode change protocols to switch from one mode to another: the Suspend/resume protocol, the Abort protocol, and the Complete protocol. These protocols increase the flexibility of the system, allowing users to select the way they want to switch to a new mode. The implementation of MMHSF is tested and evaluated on an AVR-based 32 bit board EVK1100 with an AVR32UC3A0512 micro-controller. We have tested the behavior of each system mode and for each mode change protocol. We also provide the results for the performance measures of all mode change protocols in the thesis. RESUMEN Los conmutadores de modo son usados para particionar el comportamiento del sistema en diferentes modos, reduciendo así la complejidad de grandes sistemas empotrados. Estos sistemas tienen multiples modos de operación, cada uno de ellos correspondiente a distintos escenarios y para distintas aplicaciones; son llamados Sistemas Multimodales (o en inglés “Multi-Mode Systems” o MMS). Normalmente cada modo ejecuta una parte de código distinto. En un momento dado el sistema, que está en un modo concreto, puede ser cambiado a otro modo distinto como resultado de alguna condicion impuesta previamente. Los mecanismos de cambio de modo (o protocolos de cambio de modo) son usados para mover el sistema de un modo a otro durante el tiempo de ejecución. En este trabajo se ha usado un modelo de sistema operativo para implementar un sistema multimodo llamado MMHSF, siglas en inglés correspondientes a (Multi-Mode Hierarchical Scheduling Framework). Este sistema está basado en el HSF (Hierarchical Scheduling Framework), un modelo de sistema operativo con jerarquía de dos niveles, implementado en un sistema operativo en tiempo real de libre distribución llamado FreeRTOS, capaz de permitir el aislamiento temporal entre componentes. La principal contribución de este trabajo es la ampliación del HSF convirtiendolo en un sistema multimodo realizando los cambios mínimos necesarios sobre el sistema operativo FreeRTOS y la implementación ya existente del HSF. Esta implementación usa un sistema de planificación de prioridad fija para ambos niveles de jerarquía, ocupando el tiempo entre tareas con un “modo reposo”. Además el sistema es capaz de cambiar de un modo a otro en tiempo de ejecución. Cada subsistema y tarea son capaces de tener distintos atributos de tiempo (prioridad, periodo y tiempo de ejecución) en función del modo. Bajo una demanda de cambio de modo (Mode Change Request MCR) se puede variar el set de tareas en ejecución, así como los atributos de los servidores y las tareas. Un protocolo de cambio de modo espeficica precisamente cómo será cambiado el sistema de un modo a otro. Sin embargo una aplicación puede requerir no solo un cambio de modo, sino que lo haga de una forma especifica. Por ejemplo, el cambio de modo de “normal” a “apagado” puede permitir a las tareas en ejecución ser finalizadas antes de que se complete la transición, pero sin embargo el cambio de “normal” a “emergencia” puede requerir abortar todas las tareas instantaneamente. En este trabajo ambas características, tanto el modo como el protocolo de cambio, pueden ser cambiadas en tiempo de ejecución, pero deben ser previamente definidas por el desarrollador. Han sido definidos tres protocolos de cambios: el protocolo “suspender/continuar”, protocolo “abortar” y el protocolo “completar”. Estos protocolos incrementan la flexibilidad del sistema, permitiendo al usuario seleccionar de que forma quieren cambiar hacia el nuevo modo. La implementación del MMHSF ha sido testada y evaluada en una placa AVR EVK1100, con un micro-controlador AVR32UC3A0. Se ha comprobado el comportamiento de los distintos modos para los distintos protocolos, definidos previamente. Como resultado se proporcionan las medidades de rendimiento de los distintos protocolos de cambio de modo.

Real-time scheduling of embedded applications on multi-core platforms

For the past several decades, we have experienced the tremendous growth, in both scale and scope, of real-time embedded systems, thanks largely to the advances in IC technology. However, the traditional approach to get performance boost by increasing CPU frequency has been a way of past. Researchers from both industry and academia are turning their focus to multi-core architectures for continuous improvement of computing performance. In our research, we seek to develop efficient scheduling algorithms and analysis methods in the design of real-time embedded systems on multi-core platforms. Real-time systems are the ones with the response time as critical as the logical correctness of computational results. In addition, a variety of stringent constraints such as power/energy consumption, peak temperature and reliability are also imposed to these systems. Therefore, real-time scheduling plays a critical role in design of such computing systems at the system level. We started our research by addressing timing constraints for real-time applications on multi-core platforms, and developed both partitioned and semi-partitioned scheduling algorithms to schedule fixed priority, periodic, and hard real-time tasks on multi-core platforms. Then we extended our research by taking temperature constraints into consideration. We developed a closed-form solution to capture temperature dynamics for a given periodic voltage schedule on multi-core platforms, and also developed three methods to check the feasibility of a periodic real-time schedule under peak temperature constraint. We further extended our research by incorporating the power/energy constraint with thermal awareness into our research problem. We investigated the energy estimation problem on multi-core platforms, and developed a computation efficient method to calculate the energy consumption for a given voltage schedule on a multi-core platform. In this dissertation, we present our research in details and demonstrate the effectiveness and efficiency of our approaches with extensive experimental results.

The utilization bound of uniprocessor preemptive slack-monotonic scheduling is 50%

Consider the problem of scheduling a set of sporadically arriving implicit-deadline tasks to meet deadlines on a uniprocessor. Static-priority scheduling is considered using the slack-monotonic priority-assignment scheme. We prove that its utilization bound is 50%.

Efficient schedulability tests for real-time embedded systems with urgent routines

Task scheduling is one of the key mechanisms to ensure timeliness in embedded real-time systems. Such systems have often the need to execute not only application tasks but also some urgent routines (e.g. error-detection actions, consistency checkers, interrupt handlers) with minimum latency. Although fixed-priority schedulers such as Rate-Monotonic (RM) are in line with this need, they usually make a low processor utilization available to the system. Moreover, this availability usually decreases with the number of considered tasks. If dynamic-priority schedulers such as Earliest Deadline First (EDF) are applied instead, high system utilization can be guaranteed but the minimum latency for executing urgent routines may not be ensured. In this paper we describe a scheduling model according to which urgent routines are executed at the highest priority level and all other system tasks are scheduled by EDF. We show that the guaranteed processor utilization for the assumed scheduling model is at least as high as the one provided by RM for two tasks, namely 2(2√−1). Seven polynomial time tests for checking the system timeliness are derived and proved correct. The proposed tests are compared against each other and to an exact but exponential running time test.

Holistic Analysis for Fork-Join Distributed Tasks supported by the FTT-SE Protocol

11th IEEE World Conference on Factory Communication Systems (WFCS 2015). 27 to 29, May, 2015, TII-SS-2: Scheduling and Performance Analysis. Palma de Mallorca, Spain.

An adaptive scheduling scheme for fair bandwidth allocation

Class-based service differentiation is provided in DiffServ networks. However, this differentiation will be disordered under dynamic traffic loads due to the fixed weighted scheduling. An adaptive weighted scheduling scheme is proposed in this paper to achieve fair bandwidth allocation among different service classes. In this scheme, the number of active flows and the subscribed bandwidth are estimated based on the measurement of local queue metrics, then the scheduling weights of each service class are adjusted for the per-flow fairness of excess bandwidth allocation. This adaptive scheme can be combined with any weighted scheduling algorithm. Simulation results show that, comparing with fixed weighted scheduling, it effectively improve the fairness of excess bandwidth allocation.

'Bayesian Optimisation for Nurse Scheduling'

A Bayesian optimisation algorithm for a nurse scheduling problem is presented, which involves choosing a suitable scheduling rule from a set for each nurse's assignment. When a human scheduler works, he normally builds a schedule systematically following a set of rules. After much practice, the scheduler gradually masters the knowledge of which solution parts go well with others. He can identify good parts and is aware of the solution quality even if the scheduling process is not yet completed, thus having the ability to finish a schedule by using flexible, rather than fixed, rules. In this paper, we design a more human-like scheduling algorithm, by using a Bayesian optimisation algorithm to implement explicit learning from past solutions. A nurse scheduling problem from a UK hospital is used for testing. Unlike our previous work that used Genetic Algorithms to implement implicit learning [1], the learning in the proposed algorithm is explicit, i.e. we identify and mix building blocks directly. The Bayesian optimisation algorithm is applied to implement such explicit learning by building a Bayesian network of the joint distribution of solutions. The conditional probability of each variable in the network is computed according to an initial set of promising solutions. Subsequently, each new instance for each variable is generated by using the corresponding conditional probabilities, until all variables have been generated, i.e. in our case, new rule strings have been obtained. Sets of rule strings are generated in this way, some of which will replace previous strings based on fitness. If stopping conditions are not met, the conditional probabilities for all nodes in the Bayesian network are updated again using the current set of promising rule strings. For clarity, consider the following toy example of scheduling five nurses with two rules (1: random allocation, 2: allocate nurse to low-cost shifts). In the beginning of the search, the probabilities of choosing rule 1 or 2 for each nurse is equal, i.e. 50%. After a few iterations, due to the selection pressure and reinforcement learning, we experience two solution pathways: Because pure low-cost or random allocation produces low quality solutions, either rule 1 is used for the first 2-3 nurses and rule 2 on remainder or vice versa. In essence, Bayesian network learns 'use rule 2 after 2-3x using rule 1' or vice versa. It should be noted that for our and most other scheduling problems, the structure of the network model is known and all variables are fully observed. In this case, the goal of learning is to find the rule values that maximize the likelihood of the training data. Thus, learning can amount to 'counting' in the case of multinomial distributions. For our problem, we use our rules: Random, Cheapest Cost, Best Cover and Balance of Cost and Cover. In more detail, the steps of our Bayesian optimisation algorithm for nurse scheduling are: 1. Set t = 0, and generate an initial population P(0) at random; 2. Use roulette-wheel selection to choose a set of promising rule strings S(t) from P(t); 3. Compute conditional probabilities of each node according to this set of promising solutions; 4. Assign each nurse using roulette-wheel selection based on the rules' conditional probabilities. A set of new rule strings O(t) will be generated in this way; 5. Create a new population P(t+1) by replacing some rule strings from P(t) with O(t), and set t = t+1; 6. If the termination conditions are not met (we use 2000 generations), go to step 2. Computational results from 52 real data instances demonstrate the success of this approach. They also suggest that the learning mechanism in the proposed approach might be suitable for other scheduling problems. Another direction for further research is to see if there is a good constructing sequence for individual data instances, given a fixed nurse scheduling order. If so, the good patterns could be recognized and then extracted as new domain knowledge. Thus, by using this extracted knowledge, we can assign specific rules to the corresponding nurses beforehand, and only schedule the remaining nurses with all available rules, making it possible to reduce the solution space. Acknowledgements The work was funded by the UK Government's major funding agency, Engineering and Physical Sciences Research Council (EPSRC), under grand GR/R92899/01. References [1] Aickelin U, "An Indirect Genetic Algorithm for Set Covering Problems", Journal of the Operational Research Society, 53(10): 1118-1126,

'Bayesian Optimisation for Nurse Scheduling'

A Bayesian optimisation algorithm for a nurse scheduling problem is presented, which involves choosing a suitable scheduling rule from a set for each nurse's assignment. When a human scheduler works, he normally builds a schedule systematically following a set of rules. After much practice, the scheduler gradually masters the knowledge of which solution parts go well with others. He can identify good parts and is aware of the solution quality even if the scheduling process is not yet completed, thus having the ability to finish a schedule by using flexible, rather than fixed, rules. In this paper, we design a more human-like scheduling algorithm, by using a Bayesian optimisation algorithm to implement explicit learning from past solutions. A nurse scheduling problem from a UK hospital is used for testing. Unlike our previous work that used Genetic Algorithms to implement implicit learning [1], the learning in the proposed algorithm is explicit, i.e. we identify and mix building blocks directly. The Bayesian optimisation algorithm is applied to implement such explicit learning by building a Bayesian network of the joint distribution of solutions. The conditional probability of each variable in the network is computed according to an initial set of promising solutions. Subsequently, each new instance for each variable is generated by using the corresponding conditional probabilities, until all variables have been generated, i.e. in our case, new rule strings have been obtained. Sets of rule strings are generated in this way, some of which will replace previous strings based on fitness. If stopping conditions are not met, the conditional probabilities for all nodes in the Bayesian network are updated again using the current set of promising rule strings. For clarity, consider the following toy example of scheduling five nurses with two rules (1: random allocation, 2: allocate nurse to low-cost shifts). In the beginning of the search, the probabilities of choosing rule 1 or 2 for each nurse is equal, i.e. 50%. After a few iterations, due to the selection pressure and reinforcement learning, we experience two solution pathways: Because pure low-cost or random allocation produces low quality solutions, either rule 1 is used for the first 2-3 nurses and rule 2 on remainder or vice versa. In essence, Bayesian network learns 'use rule 2 after 2-3x using rule 1' or vice versa. It should be noted that for our and most other scheduling problems, the structure of the network model is known and all variables are fully observed. In this case, the goal of learning is to find the rule values that maximize the likelihood of the training data. Thus, learning can amount to 'counting' in the case of multinomial distributions. For our problem, we use our rules: Random, Cheapest Cost, Best Cover and Balance of Cost and Cover. In more detail, the steps of our Bayesian optimisation algorithm for nurse scheduling are: 1. Set t = 0, and generate an initial population P(0) at random; 2. Use roulette-wheel selection to choose a set of promising rule strings S(t) from P(t); 3. Compute conditional probabilities of each node according to this set of promising solutions; 4. Assign each nurse using roulette-wheel selection based on the rules' conditional probabilities. A set of new rule strings O(t) will be generated in this way; 5. Create a new population P(t+1) by replacing some rule strings from P(t) with O(t), and set t = t+1; 6. If the termination conditions are not met (we use 2000 generations), go to step 2. Computational results from 52 real data instances demonstrate the success of this approach. They also suggest that the learning mechanism in the proposed approach might be suitable for other scheduling problems. Another direction for further research is to see if there is a good constructing sequence for individual data instances, given a fixed nurse scheduling order. If so, the good patterns could be recognized and then extracted as new domain knowledge. Thus, by using this extracted knowledge, we can assign specific rules to the corresponding nurses beforehand, and only schedule the remaining nurses with all available rules, making it possible to reduce the solution space. Acknowledgements The work was funded by the UK Government's major funding agency, Engineering and Physical Sciences Research Council (EPSRC), under grand GR/R92899/01. References [1] Aickelin U, "An Indirect Genetic Algorithm for Set Covering Problems", Journal of the Operational Research Society, 53(10): 1118-1126,

Slotted wiDom: schedulability analysis and its experimental validation

WiDom is a wireless prioritized medium access control protocol which offers a very large number of priority levels. Hence, it brings the potential to employ non-preemptive static-priority scheduling and schedulability analysis for a wireless channel assuming that the overhead of WiDom is modeled properly. One schedulability analysis for WiDom has already been proposed but recent research has created a new version of WiDom (we call it: Slotted WiDom) with lower overhead and for this version of WiDom no schedulability analysis exists. In this paper we propose a new schedulability analysis for slotted WiDom and extend it to work also for message streams with release jitter. We have performed experiments with an implementation of slotted WiDom on a real-world platform (MicaZ). We find that for each message stream, the maximum observed response time never exceeds the calculated response time and hence this corroborates our belief that our new scheduling theory is applicable in practice.

New schedulability analysis for WiDom

WiDom is a wireless prioritized medium access control (MAC) protocol which offers a very large number of priority levels. Hence, it brings the potential for employing non-preemptive static-priority scheduling and schedulability analysis for a wireless channel assuming that the overhead of WiDom is modeled properly. One schedulability analysis for WiDom has already been proposed but recent research has created a new version of WiDom with lower overhead (we call it: WiDom with a master node) and for this version of WiDom no schedulability analysis exists. Also, common to the previously proposed schedulability analyses for WiDom is that they cannot analyze message streams with release jitter. Therefore, in this paper we propose a new schedulability analysis for WiDom (with a master node). We also extend the WiDom analyses (with and without master node) to work also for message streams with release jitter.