Round-robin Scheduling Policy Research Articles

Average Age of Information in Update Systems With Active Sources and Packet Delivery Errors

This letter studies the “age of information” (AoI) in a multi-source status update system where N active sources each send updates of their time-varying process to a monitor through a server with packet delivery errors. We analyze the average AoI for stationary randomized and round-robin scheduling policies. For both of these scheduling policies, we further analyze the effect of packet retransmission policies, i.e., retransmission without resampling, retransmission with resampling, or no retransmission, when errors occur. Expressions for the average AoI are derived for each case. It is shown that the round-robin schedule policy in conjunction with retransmission with resampling when errors occur achieves the lowest average AoI among the considered cases. For stationary randomized schedules with equiprobable source selection, it is further shown that the average AoI gap to round-robin schedules with the same packet management policy scales as O(N ). Finally, for stationary randomized policies, the optimal source selection probabilities that minimize a weighted sum average AoI metric are derived.

An SDN-enhanced load-balancing technique in the cloud system

The vast majority of Web services and sites are hosted in various kinds of cloud services, and ordering some level of quality of service (QoS) in such systems requires effective load-balancing policies that choose among multiple clouds. Recently, software-defined networking (SDN) is one of the most promising solutions for load balancing in cloud data center. SDN is characterized by its two distinguished features, including decoupling the control plane from the data plane and providing programmability for network application development. By using these technologies, SDN and cloud computing can improve cloud reliability, manageability, scalability and controllability. SDN-based cloud is a new type cloud in which SDN technology is used to acquire control on network infrastructure and to provide networking-as-a-service (NaaS) in cloud computing environments. In this paper, we introduce an SDN-enhanced Inter cloud Manager (S-ICM) that allocates network flows in the cloud environment. S-ICM consists of two main parts, monitoring and decision making. For monitoring, S-ICM uses SDN control message that observes and collects data, and decision-making is based on the measured network delay of packets. Measurements are used to compare S-ICM with a round robin (RR) allocation of jobs between clouds which spreads the workload equitably, and with a honeybee foraging algorithm (HFA). We see that S-ICM is better at avoiding system saturation than HFA and RR under heavy load formula using RR job scheduler. Measurements are also used to evaluate whether a simple queueing formula can be used to predict system performance for several clouds being operated under an RR scheduling policy, and show the validity of the theoretical approximation.

Delay-Privacy Tradeoff in the Design of Scheduling Policies

Traditionally, scheduling policies have been optimized to perform well on metrics, such as throughput, delay, and fairness. In the context of shared event schedulers, where a common processor is shared among multiple users, one also has to consider the privacy offered by the scheduling policy. The privacy offered by a scheduling policy measures how much information about the usage pattern of one user of the system can be learned by another as a consequence of sharing the scheduler. We introduced an estimation error-based metric to quantify this privacy. We showed that the most commonly deployed scheduling policy, the first-come-first-served offers very little privacy to its users. We also proposed a parametric nonwork conserving policy, which traded off delay for improved privacy. In this paper, we ask the question, is a tradeoff between delay and privacy fundamental to the design to scheduling policies? In particular, is there a work conserving, possibly randomized, and scheduling policy that scores high on the privacy metric? Answering the first question, we show that there does exist a fundamental limit on the privacy performance of a work-conserving scheduling policy. We quantify this limit. Furthermore, answering the second question, we demonstrate that the round-robin scheduling policy (deterministic policy) is privacy optimal within the class of work-conserving policies.

A SLA-based cloud resource provisioning optimisation mechanism for multi-tenant applications

As cloud resources exhibit the characteristics of dynamic and scalability, an increasing number of service providers deploy their multi-tenant applications to Cloud. For cloud service providers, how to arrange the allocation of application resources reasonably, in order to not only satisfy the service level agreements (SLAs) of diverse tenants, but also maximise their resource utilisation, has become a major challenge. In this paper, we propose a SLA-based resource provisioning optimisation mechanism for multi-tenant applications. Based on the invoking relationships among services and the workload of service instances, the mechanism divides the tenants' SLA into SLA constraints, and maps each SLA constraint to the corresponding service. In addition, the mechanism uses an optimised round-robin scheduling policy to improve the resource utilisation of each service instance, and employs a defragmentation optimisation algorithm to control the scale of service instances and increase the overall resource utilisation. The experimental results show that the proposed mechanism is feasible and effective.

Enhancing the real-time capabilities of the Linux kernel

The mainline Linux Kernel is not designed for hard real-time systems; it only fits the requirements of soft real-time systems. In recent years, a kernel developer community has been working on the PREEMPT-RT patch. This patch (that aims to get a fully preemptible kernel) adds some real-time capabilities to the Linux kernel. However, in terms of scheduling policies, the real-time scheduling class of Linux is limited to the First-In-First-Out (SCHED_FIFO) and Round-Robin (SCHED_RR) scheduling policies. These scheduling policies are however quite limited in terms of real-time performance. Therefore, in this paper, we report one important contribution for adding more advanced real-time capabilities to the Linux Kernel. Specifically, we describe modifications to the (PREEMPT-RT patched) Linux kernel to support real-time slot-based task-splitting scheduling algorithms. Our preliminary evaluation shows that our implementation exhibits a real-time performance that is superior to the scheduling policies provided by the current version of PREMPT-RT. This is a significant add-on to a widely adopted operating system.

Analyzing Weighted Round Robin policies with a stochastic comparison approach

We study queuing delays for Weighted Round Robin ( WRR) scheduling policies. The delay characteristics of these policies can be estimated through worst-case bounds as other fair queuing scheduling policies. However, these deterministic bounds are in general not accurate and they require that the input process has been shaped. Under Markovian arrival hypothesis, the delay of WRR policies can be evaluated through Markovian numerical analysis. Nevertheless, this analysis is limited to small parameter sizes because of the state space explosion problem. We propose to apply the stochastic comparison approach. We build a bounding model based on the aggregation of sessions, and we show that under the same arrivals, the packet delays in the bounding model are larger in the ⩽ st stochastic order sense than the packet delays in the original model. Due to this aggregation, the state space size is drastically reduced and we are able to provide stochastic delay bounds for WRR policies. We discuss the accuracy of the proposed bounds through numerical examples.

Timing analysis of compound scheduling policies: application to posix1003.1b

The analysis of fixed priority preemptive scheduling has been extended in various ways to improve its usefulness for the design of real-time systems. In this paper, we define the layered preemptive priority scheduling policy which generalizes fixed preemptive priorities by combination with other policies in a layered structure. In particular, the combination with the Round Robin scheduling policy is studied. Its compliance with Posix 1003.1b requirements is shown and its timing analysis is provided. For this purpose and as a basis for the analysis of other policies, the concept of majorizing work arrival function, is introduced to synthesize essential ideas used in existing analysis of the fixed preemptive priority policy.If critical resources are protected by semaphores, the Priority Ceiling Protocol (PCP) can be used under fixed preemptive priorities to control resulting priority inversions. An extension of the PCP is proposed for Round Robin, to allow a global control of priority inversions under the layered priority policy and to prevent deadlocks. The initial timing analysis is extended to account for the effects of the protocol. The results are illustrated by a small test case.

Distributed scheduling of flexible manufacturing systems: stability and performance

We consider a manufacturing system producing several part-types on several machines. Raw parts are input to the system. Each unit of a given part-type requires a predetermined processing time at each of several machines, in a given order. A setup time is required whenever a machine switches from processing one part-type to another. For a single machine system with constant demand rates, we present a class of generalized round-robin scheduling policies for which the buffer level trajectory of each part-type converges to a steady state level. Furthermore, for all small initial conditions, we show that these policies can be Pareto-efficient with respect to the buffer sizes required. Allowing the input streams to have some burstiness, we derive upper bounds on the buffer levels for small initial conditions. For non-acyclic systems, we consider a class of policies which are stable for all inputs with bounded burstiness. We show how to employ system elements, called regulators, to stabilize systems. Using the bounds for the single machine case, we analyze the performance of regulated systems implementing generalized round-robin scheduling policies. >

Scheduler for realtime process control

Microcomputers are increasingly being used in various process control systems. A recent trend is to use them in multiprogrammed realtime environments rather than as dedicated controllers. The paper describes the design of a scheduler for realtime process control using microcomputers. For simplicity, a static-priority-based round-robin scheduling policy is used whereby the working time slice can be varied for the different processes. The memory size for the process scheduler is estimated at approximately 1 kbyte using the Intel 8085 instruction set.

A Real-Time Process Scheduler for a Ring-Type Microcomputer Network

This paper describes the design of a real-time process control scheduler for microcomputers in a distributed process control scheme with a ring-type network configuration. For simplicity, static priority based round-robin scheduling policy is used where working time-slice, for different processes can be variable. The memory size for the process scheduler modules is estimated at about 2.5 Kbytes using an Intel 8085 instruction set and a RAM requirement of about 1 Kbyte.