High-level Scheduling Research Articles

A framework for integrated resource planning in surgical clinics

The problem under study is based on the challenges faced by the Orthopaedic Clinic at St. Olav’s Hospital in Trondheim, Norway. Variations in demand and supply cause fluctuating waiting lists, and it is challenging to level the activities between the clinic’s two units, the outpatient clinic and the operating theater, to obtain short waiting times for all activities. Based on these challenges, we describe and present a planning problem referred to as the Long-term Master Scheduling Problem (LMSP), where the objective is to construct an integrated Long-term Master Schedule (LMS) that facilitates short waiting times in both units. The LMS can be separated into two schedules, one cyclic high-level schedule, and one non-cyclic low-level schedule. The demand for outpatient clinic consultations and surgeries is stochastic, as are the waiting lists. To account for this, we propose a planning framework consisting of an optimization model to solve the LMSP, and a two-level planning procedure. In the planning procedure, we first solve the LMSP to construct the LMS for the upcoming planning horizon. Then, to adjust to the fluctuating waiting lists, we periodically refine the low-level schedule by solving a constrained LMSP. We also develop a simulation-based evaluation procedure to evaluate the planning framework in a real-life setting and use this to investigate different planning strategies. We find that imposing flexible, dynamic and agile planning strategies improve waiting time outcomes and patient throughput. Furthermore, combining the strategies yields additive improvements.

Model predictive approaches for cost-efficient building climate control with seasonal energy storage

Due to the increasing share of fluctuating renewable energy, a seasonal energy storage with a high capacity is needed, which stores the summer surplus electricity to satisfy the winter heating demand.It is assumed that the control integration of a novel thermochemical seasonal energy storage concept into a building energy system can meet this requirement and save operating costs. For this purpose, a model-predictive control concept with a prediction horizon beyond that of public weather forecasts is crucial.A state-based model is developed consisting of a building, a water buffer and a heat supply. To assess the effects of seasonal storage, different heat supply configurations are considered. On this basis, two Model Predictive Control (MPC) concepts are designed to efficiently operate the system over one year. Since public weather forecasts are reliable in the time range of several days, test reference year data are used to approximate the weather forecast beyond the public forecast period. Additionally, the control hierarchy is comprised of a superordinate optimal generation scheduling (OGS) and a subordinate MPC. The concepts follow the scheduled long-term lime storage trajectory and realise possible short-term yields based on the current public forecast. The trajectory tracking is formulated either in the objective function or the constraints.The integration of the novel lime storage module into the heat supply of a building allows a reduction of operating costs of 18% in the realistic scenario and up to 80% in case of highly fluctuating electricity prices. This reduction potential is fully exploited by the developed control approaches, but it is very sensitive to the change of the controller parameters, the fluctuation of the electricity price and the weather data. Moreover, by applying the objective-based reference tracking approach, the higher-level scheduling hierarchy could be avoided.Through the results of this work, it is confirmed that the integration of a seasonal energy storage system can greatly reduce annual operating costs and that it is not crucial to apply a prediction horizon beyond that of public forecasts.

Verified tensor-program optimization via high-level scheduling rewrites

We present a lightweight Coq framework for optimizing tensor kernels written in a pure, functional array language. Optimizations rely on user scheduling using series of verified, semantics-preserving rewrites. Unusually for compilation targeting imperative code with arrays and nested loops, all rewrites are source-to-source within a purely functional language. Our language comprises a set of core constructs for expressing high-level computation detail and a set of what we call reshape operators, which can be derived from core constructs but trigger low-level decisions about storage patterns and ordering. We demonstrate that not only is this system capable of deriving the optimizations of existing state-of-the-art languages like Halide and generating comparably performant code, it is also able to schedule a family of useful program transformations beyond what is reachable in Halide.

Intelligent scheduling and flexible operation for the commercial-scale coal-fired power plant integrated with post-combustion carbon capture

This paper develops a two-level control structure for commercial-scale coal-fired power plant integrated with post-combustion carbon capture (CFPP-PCC). In high-level scheduler, an economic objective function with the consideration of grid demand, price conditions and system constraints is developed. Deep belief network is utilized to develop the steady-state model for CFPP-PCC process. Intelligent scheduling of electricity generation and capture level is then implemented using Bayesian optimization. Given the optimal setpoints, model predictive controller (MPC) is applied to achieve flexible operation in the low-level supervisory control. The simulation results show that the proposed control structure specifies an economically attractive operation profile for the CFPP-PCC process.

The plan forward for EU DEMO

In the preparation for the Concept Design Phase and a new European Framework Programme starting in 2021, EUROfusion developed a DEMO project charter, consisting of updates to the project governance and the high level objectives, schedule and scope of work. This proposal, together with the more detailed proposals covering the scope of system design and technology R&D across the breadth of the DEMO project, forms the plan forward for DEMO and was submitted as input for approval to the first DEMO gate review (G1). Following the successful completion of the gate and the associated recommendations provided, the plan forward was modified accordingly. This paper details the principle elements of the implementation and organisation of the DEMO design and R&D activities in the Concept Design Phase.

ALGORITHMIZATION AND MOTION PROGRAMMING OF THE VACUUM CLEANER-ROBOT WITH PATH CORRECTION

The reliability of navigation in a dynamic or unknown environment depends on the ability of robot vacuum cleaners to move between unknown obstacles without collisions and rapid response to uncertainty. Problems associated with behavioral navigation systems are behavior coordination or choice of actions. Multiple behavior can simultaneously generate multiple command outputs, which can lead to the movement of the robot in unforeseen directions or to the complete failure of the system. Reliable system performance depends on deciding how to integrate high-level scheduling and low-level execution behavior, which behavior should be activated, and how the source commands should be combined into a single robot management command. The choice of fuzzy logic system is justified by the fact that it is ideal for navigation, as for a robot vacuum cleaner itself, and any other mobile robot that requires navigation in space: it is easy to use andimplement in software code. In essence, it resembles the use of the function IF… Then, but with much greater opportunities for the distribution and use of input values. This in turn allows for smooth adjustment of the output values, in our case – the angle of rotation, as well as speed. This research uses the CENTER OF GRAVITY (COG) method, which provides a clear value based on the center of gravity of a fuzzy set and is considered the most accurate of all defasification methods.This article presents the results of the development of an environment for simulating the movement of a robot vacuum cleaner, which allows one to estimate the differences between the applied algorithms for using fuzzy logic and the reliability of the chosen method. It is also possible to evaluate the effectiveness of the embedded logic for a path-choosing. In the simulation, it is possible to import a room layout, which gives an idea of the need to implement complex logic.

Hybrid multi-mode simulation and optimization for subarea scheduling in heavy industrial construction

Scheduling heavy industrial projects is challenging due to their size and complexity. Three-dimensional space and its discipline-specific tasks are an important asset in congested industrial sites, yet are not easily scheduled early in planning when detailed information is lacking. Although recent approaches have focused on enhancing space and crew management, effective scheduling of large-scale industrial projects with several interdependent constraints remains difficult. A comprehensive, hybrid planning framework capable of developing an efficient, high-level schedule accounting for uncertainties in bidding and execution is proposed. The framework employs simulation and optimization tools to determine best durations/modes in multi-mode activity networks and uses dynamic data adoption to continuously plan during project progression. The proposed framework uses historical information and expert opinion to automatically create a near-optimized schedule while considering various uncertainties. The proposed approach is tested and demonstrated with an illustrative example and an actual case study of industrial construction in Alberta, Canada.

Improving Performance of Tape Restore Request Scheduling in the Storage System dCache

Given the anticipated increase in the amount of scientific data, it is widely accepted that primarily disk based storage will become prohibitively expensive. Tape based storage, on the other hand, provides a viable and affordable solution for the ever increasing demand for storage space. Coupled with a disk caching layer that temporarily holds a small fraction of the total data volume to allow for low latency access, it turns tape based systems into active archival storage (write once, read many) that imposes additional demands on data flow optimization compared to traditional backup setups (write once, read never). In order to preserve the lifetime of tapes and minimize the inherently higher access latency, different tape usage strategies are being evaluated. As an important disk storage system for scientific data that transparently handles tape access, dCache is making efforts to evaluate its recall optimization potential and is introducing a proof-of-concept, high-level stage request scheduling component within its SRM implementation.

Integrating discrete event simulation and genetic algorithm optimization for bridge maintenance planning

To minimize agency and user costs in a bridge repair project, a bridge maintenance manager should develop an appropriate project schedule considering real-world constraints such as resource limitations (e.g., workspace and crew). This paper presents a new framework called Simulation-based Bridge Maintenance Optimization (SiBMO) by integrating Genetic Algorithm (GA) and Discrete Event Simulation (DES) to identify the optimum maintenance plan taking into account crew limitations. The framework optimizes the sequence of repair-activities in the repair interventions considering workspace limitations and predecessor relationships. SiBMO also develops a high-level schedule of the interventions regarding the project calendar and the Traffic Management Plan (TMP). The Bridge Information Model (BrIM) based user interface developed in this study visualizes the high-level schedule. The results of applying SiBMO on a real case study demonstrates its capability in finding the optimum maintenance plan, its efficiency in optimizing the high-level schedule, and its accuracy in estimating user costs.

An automatic solution to make HTCondor more stable and easier

HTCondor has been widely adopted by HEP clusters to provide high-level scheduling performance. Unlike other schedulers, HTCondor provides loose management of the worker nodes. We developed a maintenance automation tool called “HTCondor MAT” that focuses on dynamic resource management and automatic error handling. A central database records all worker node information, which is sent to the worker node for the startd configuration. If an error happens for the worker node, the node information stored in the database is updated and the worker node is reconfigured with the new node information. The new configuration stops the startd from accepting error-related jobs until the worker node recovers. The MAT has been deployed in the IHEP HTC cluster to provide a central way to manage the worker nodes and remove the impacts of errors on the worker nodes automatically.

Modelling and cooling power control of a TES-backed-up vapour-compression refrigeration system

This work addresses the modelling, power control, and optimization of a thermal energy storage (TES) system combined with a vapour-compression refrigeration facility based on phase change materials (PCM). Given a novel design of a PCM-based TES tank and its interconnection with an existing refrigeration system, the joint dynamic modelling is first studied, exploring the different time scales that coexist at the interconnected system. Diverse operating modes are defined, according to the intended use of the TES tank as a cold-energy buffer to decouple cooling demand and production, whereas the static characteristic and power limits are calculated and show the high coupling between the main cooling powers involved (TES charging/discharging power, and direct power production at the evaporator). In this light, a decoupling control strategy is proposed, where the low-level controllers are simply PI regulators and the refrigerant/secondary mass flows are considered as virtual manipulated variables, applying a feedforward-based cascade strategy. The control performance is evaluated through a thorough simulation that includes all operating modes, where the reference tracking is shown to be fast and reliable enough to address high-level scheduling strategies, where the references on the main cooling powers are intended to be imposed considering economic and efficiency criteria.

SAASFEE

Across many fields of science, primary data sets like sensor read-outs, time series, and genomic sequences are analyzed by complex chains of specialized tools and scripts exchanging intermediate results in domain-specific file formats. Scientific workflow management systems (SWfMSs) support the development and execution of these tool chains by providing workflow specification languages, graphical editors, fault-tolerant execution engines, etc. However, many SWfMSs are not prepared to handle large data sets because of inadequate support for distributed computing. On the other hand, most SWfMSs that do support distributed computing only allow static task execution orders. We present SAASFEE, a SWfMS which runs arbitrarily complex workflows on Hadoop YARN. Workflows are specified in Cuneiform, a functional workflow language focusing on parallelization and easy integration of existing software. Cuneiform workflows are executed on Hi-WAY, a higher-level scheduler for running workflows on YARN. Distinct features of SAASFEE are the ability to execute iterative workflows, an adaptive task scheduler, re-executable provenance traces, and compatibility to selected other workflow systems. In the demonstration, we present all components of SAASFEE using real-life workflows from the field of genomics.

DNS Based Approach Load Balancing In Distributed Web Server System

In this paper, we consider the high level scheduling and load sharing properties offered by the Domain Name System, as implemented in popular DNS software packages. We explore the performance and capabilities of high-level DNS-based load balancing, where we draw special attention to the choice of caching policy (time-to-live) for DNS data. Further, we analyze the use of a database-supported DNS service for allowing highly dynamical query responses, and show that this approach has both potentially negative (single point of failure) and positive (improved balancing flexibility) properties. The objective of this paper is we discuss the performance requirements of the DNS, and argue that the robustness and performance of the DNS could be improved and also discuss the impact of TTL on response time in DNS

A high-speed unified hardware architecture for 128 and 256-bit security levels of AES and the SHA-3 candidate Grøstl

One of the five final SHA-3 candidates, Grøstl, has been inspired by the Advanced Encryption Standard. This unique feature can be exploited in a large variety of practical applications. In order to have a better picture of the Grøstl–AES computational efficiency (high-level scheduling, internal pipelining, resource sharing, etc.), we designed a high-speed coprocessor for the Grøstl-based HMAC and AES in the counter mode. This coprocessor offers high-speed computations of both authentication and encryption/decryption with relatively small penalty in terms of area and speed when compared to the authentication (original Grøstl circuitry) functionality only. From our perspective, the main advantage of Grøstl over other finalists is the fact that its hardware architecture naturally accommodates AES at the cost of a small area overhead.

High-Level Scheduling of Energy Optimal Trajectories

The reduction of energy consumption is today addressed with great effort in manufacturing industry. In this paper, we improve upon a previously presented method for robotic system scheduling. By applying dynamic programming to existing trajectories, we generate new energy optimal trajectories that follow the same path but in a different execution time frame. With this new method, it is possible to solve the optimization problem for a range of execution times for the individual operations, based on one simulation only. The minimum energy trajectories can then be used to derive a globally energy optimal schedule. A case study of a cell comprised of four six-link manipulators is presented, in which energy optimal dynamic time scaling is compared to linear time scaling. The results show that a significant decrease in energy consumption can be achieved for any given cycle time.

CASP: a community-aware scheduling protocol

The existing resource and topology heterogeneity has divided the scheduling solutions into local schedulers and high-level schedulers (a.k.a. meta-schedulers). Although much work has been proposed to optimise job queue based scheduling, seldom has attention been put on the job sharing behaviours between decentralised distributed resource pools, which in turn raises a notable opportunity to exploit and optimise the process of job sharing between reachable grid dynamically and proactively. In our work, we introduce a novel scheduling protocol named the community-aware scheduling protocol (CASP), which dedicates to disseminate scheduling events of each participating node to as many remote nodes as possible. By means of the proposed protocol, the scheduling process of each received job consists of two phases with awareness of grid volatility. The implemented prototype and evaluated results have shown the introduced CASP is able to cooperate with a variety of local scheduling algorithms as well as diverse types of grids.

Deadlock prevention and performance oriented supervision in flexible manufacturing cells: A hierarchical approach

In this paper, a hierarchical control system is proposed for automated flexible manufacturing cells (FMCs) that operate in a job shop flow setting. The control system is made up of a higher level scheduler that optimizes the production flow within the cell, and a lower level supervisor that implements the decisions of the scheduler on the shop floor. To obtain the supervisor, a production schedule is transformed into an augmented Marked Graph (MG) model that can interact with the cell devices. Because of the flow complexities inherent in job shop systems, they are usually prone to deadlocks. Accordingly, this paper also proposes a necessary condition for deadlock occurrence in the scheduling phase. The proposed approach is validated by implementation in an experimental manufacturing cell.

User-oriented hierarchical bandwidth scheduling for ethernet passive optical networks

Ethernet passive optical networks (EPONs) are being designed to deliver different quality of service (QoS) to carry heterogeneous traffic of end users. For this purpose, hierarchical scheduling is needed for upstream bandwidth allocation, with high-level scheduling for inter-optical network unit (ONU) allocation and low-level scheduling for intra-ONU distribution. In this paper, we propose new User-oriented Hierarchical bandwidth Scheduling Algorithms (UHSAs) that support differentiated services and guaranteed fairness among end users. For inter-ONU scheduling, we adopt an improved hybrid cycle approach that separates a frame into a static part for high priority traffic and an adaptive dynamic part for low priority traffic. For intra-ONU scheduling, we propose credit-based scheduling approach to guarantee fairness among end users. To improve scheduling efficiency and lower queue management complexity, we design a novel credit-based common queue (CCQ) for each traffic class to enhance scheduling architecture and minimize average number of queues in the ONU. On the other hand, we propose a transmission priority scheme for different queue groups, which together with CCQ mechanism serves the objective of improving delay and delay variation performance of high priority traffic, guaranteeing throughput for bandwidth sensitive medium priority traffic, as well as providing fairness and throughput protection among different users. The UHSAs exhibit a feature of multiple transmission opportunities (M-opportunities) per-cycle for high priority traffic that is significant in improving delay and delay variation performance for high priority traffic as compared with previous solutions of single transmission opportunity (S-opportunity) per-cycle. Detailed simulation experiments are conducted to study the performance and validate the effectiveness of the proposed protocols.

Autonomic Data Management System in Grid Environment

In grid environment, a job may execute on a node that is geographically far away from its data files. These files are stored in heterogeneous storage systems located at geographically distributed virtual organizations. The current approach includes explicit data file transfers to execution nodes, which forces users to deal with different administrative policies at each site and various data access mechanisms on each storage system. This implies a lot of human interventions in order to develop dedicated programs and scripts for data transfers for job execution. We have developed a framework, called GRAVY, which enables transparent data access between distributed file systems irrespective of their heterogeneity. This feature enables highlevel schedulers integrated with GRAVY to control data placements like computational jobs (i.e., they can be queued, scheduled and monitored). In particular, GRAVY provides reliable and efficient data placement across multiple protocols and has a mechanism for transfer failure recovery without any human interaction.

Interrupt modeling for efficient high-level scheduler design space exploration

Single Chip Heterogeneous Multiprocessors executing a wide variety of software are increasingly common in consumer electronics. Because of the mix of real-time and best effort software across the entire chip, a key design element of these systems is the choice of scheduling strategy. Without task migration, the benefits of single chip processing cannot be fully realized. Previously, high-level modeling environments have not been capable of modeling asynchronous events such as interrupts and preemptive scheduling while preserving the performance benefits of high level simulation. This paper shows how extensions to Modeling Environment for Software and Hardware (MESH) enable precise modeling of these asynchronous events while running more than 1000 faster than cycle-accurate simulation. We discuss how we achieved this and illustrate its use in modeling preemptive scheduling. We evaluate the potential of migrating running tasks between processors to improve performance in a multimedia cell phone example. We show that by allowing schedulers to rebalance processor loads as new tasks arrive significant performance gains can be achieved over statically partitioned and dynamic scheduling approaches. In our example, we show that system response time can be improved by as much as 1.96 times when a preemptive migratory scheduler is used, despite the overhead incurred by scheduling tasks across multiple processors and transferring state during the migration of running tasks. The contribution of this work is to provide a framework for evaluating preemptive scheduling policies and task migration in a high level simulator, by combining the new ability to model interrupts with dramatically increased efficiency in the high-level modeling of scheduling and commuincation MESH already provides.