Masaki Hanada

Clustering-Based Task Scheduling in a Large Number of Heterogeneous Processors

Parallelization paradigms for effective execution in a Directed Acyclic Graph (DAG) application have been widely studied in the area of task scheduling. Schedule length can be varied depending on task assignment policies, scheduling policies, and heterogeneity in terms of each processor and each communication bandwidth in a heterogeneous system. One disadvantage of existing task scheduling algorithms is that the schedule length cannot be reduced for a data intensive application. In this paper, we propose a clustering-based task scheduling algorithm called Clustering for Minimizing the Worst Schedule Length (CMWSL) to minimize the schedule length in a large number of heterogeneous processors. First, the proposed method derives the lower bound of the total execution time for each processor by taking both the system and application characteristics into account. As a result, the number of processors used for actual execution is regulated to minimize the Worst Schedule Length (WSL). Then, the actual task assignment and task clustering are performed to minimize the schedule length until the total execution time in a task cluster exceeds the lower bound. Experimental results indicate that CMWSL outperforms both existing list-based and clustering-based task scheduling algorithms in terms of the schedule length and efficiency, especially in data-intensive applications.

ORC-GPS: Output Rate-Controlled Scheduling Policy for Delay Guarantees

Recently packet scheduling algorithms such as Packetized GPS (PGPS), worst-case fair weighted fair queueing (WF2Q) and Self-Clocked Fair Queueing (SCFQ) have been proposed in order to guarantee deterministic or statistical delay bounds. These algorithms are based on generalized processor sharing (GPS) which is an ideal scheduling algorithm based on fluid flow model in which the traffic is infinitely divisible. GPS provides a minimum guaranteed service rate for each session and tight delay bounds for leaky bucket constrained sessions. However, the delay bounds are unnecessarily large because each session is served according to its associated constant weight until the session buffer is empty. In this paper, we present a scheduling policy called output rate-controlled generalized processor sharing (ORC-GPS). ORC-GPS is a rate-based scheduling like GPS and controls the service rate to lower the delay bounds for leaky bucket constrained sessions. In numerical experiments, we compare ORC-GPS with GPS in terms of delay bounds.

Asymmetric RTS/CTS for Exposed Node Reduction in IEEE 802.11 Ad Hoc Networks

One interesting problem regarding wireless local area network (WLAN) ad-hoc networks is the effective mitigation of hidden nodes. The WLAN standard IEEE 802.11 provides request to send/clear to send (RTS/CTS) as mitigation for the hidden node problem; however, this causes the exposed node problem. The first 802.11 standard provided only two transmission rates, 1 and 2 Mbps, and control frames, such as RTS/CTS assumed to be sent at 1 Mbps. The 802.11 standard has been enhanced several times since then and now it supports multi-rate transmission up to 65 Mbps in the currently popular 802.11n (20 MHz channel, single stream with long guard interval). As a result, the difference in transmission rates and coverages between the data frame and control frame can be very large. However adjusting the RTS/CTS transmission rate to optimize network throughput has not been well investigated. In this paper, we propose a method to decrease the number of exposed nodes by increasing the RTS transmission rate to decrease RTS coverage. Our proposed method, Asymmetric Range by Multi-Rate Control (ARMRC), can decrease or even completely eliminate exposed nodes and improve the entire network throughput. Experimental results by simulation show that the network throughput in the proposed method is higher by 20% to 50% under certain conditions, and the proposed method is found to be effective in equalizing dispersion of throughput among nodes.

Output Rate-Controlled Scheduling Policy: End-to-End Delay Bounds Calculation

Recently many packet scheduling algorithms based on generalized processor sharing (GPS) and earliest deadline first (EDF) have been proposed in order to guarantee deterministic or statistical delay bounds. GPS provides a minimum guaranteed service rate for each session and tight end-to-end delay bounds for leaky bucket constrained sessions. However, the delay bounds are unnecessarily large because each session is served according to its associated constant weight until the session buffer is empty. EDF is the optimal scheduling algorithm in terms of schedulable region in a single-node network. However, using EDF to provide end-to-end delay bounds is problematic because the traffic will be distorted after traffic aggregation in a multi-node network. In this paper, we present a scheduling policy called output rate-controlled generalized processor sharing (ORC-GPS) in order to guarantee deterministic delay bounds. ORC-GPS is a rate-based scheduling like GPS and controls the service rate to lower the delay bounds for leaky bucket constrained sessions. We compare ORC-GPS with GPS and EDF in terms of delay bounds.

Ad hoc WLAN throughput improvement by reduction of RTS range

In this paper, we evaluate a novel method for reducing exposed nodes in IEEE 802.11 ad hoc WLANs using asymmetric transmission ranges for RTS and CTS frames. The RTS/CTS handshake communication control mechanism used in IEEE 802.11 networks solves the hidden node problem but causes the exposed node problem. Our proposed method uses asymmetric transmission ranges for RTS and CTS control frames to solve the exposed node problem. Simulations using the Network Simulator 2 (NS-2) show that asymmetric transmission of RTS and CTS frames improves overall network throughput compared to the standard RTS/CTS method.

Optimizing RTS/CTS to improve throughput in ad hoc WLANs

IEEE 802.11 WLANs use carrier sense multiple access with collision avoidance (CSMA/CA) to initiate the Request to Send / Clear to Send (RTS/CTS) handshaking mechanism that solves the hidden node problem. However RTS/CTS also causes the exposed node problem where a node is unnecessarily prevented from accessing the wireless channel even when such access will not disrupt another nodes ongoing transmission. In this paper, we present continuing evaluation of a method for reducing exposed nodes in 802.11 ad hoc WLANs using asymmetric transmission ranges for RTS and CTS frames. NS-2 simulations show that the proposed method improves overall network throughput in a topology scenario of a 3-D network faced with ceiling/floor obstructions.

Prior node selection for scheduling workflows in a heterogeneous system

Abstract Many workflow scheduling algorithms for heterogeneous systems have been developed to satisfy multiple requirements such as minimizing schedule length while maximizing throughput. In particular, in list-based scheduling approaches, the schedule length depends on the given nodes as well as the task allocation and ordering policies. This is because the scheduling priority is derived by averaging the execution time and communication time of the given nodes. If the set of nodes can be adjusted before the scheduling tasks, a small schedule length can be achieved. In this paper, we propose a prior node selection algorithm, called lower bound based candidate node selection (LBCNS) to select a subset of given nodes to minimize the schedule length while fairly scheduling each job. Our proposal has two approaches: (i) LBCNS_DEFAULT, which considers the job characteristics and each node’s performance, and (ii) priority-based LBCNS, which additionally takes each scheduling priority into account for a dedicated task scheduling algorithm. The experimental results of extensive simulations show that LBCNS_DEFAULT has the best fairness for scheduling multiple workflow jobs, while priority-based LBCNS achieves the minimum schedule length with the highest efficiency for a single workflow job and multiple workflow jobs.

Throughput improvement by adjusting RTS transmission range for W-LAN Ad Hoc network

The W-LAN Ad Hoc network tends to cause problems called “Hidden Node” and “Exposed Node”. RTS/CTS mechanism has been introduced to mitigate Hidden Node and most of existing researches assume that RTS and CTS are sent at the same transmission range. This paper describes a new method to improve the network throughput by adjusting the RTS transmission range. The simulation result showed that the proposed method achieved higher throughput in some degree.

Effective use of computational resources in multicore distributed systems

In the last decades, many kinds of task execution models such as grid and cloud computing have been developed. In such distributed systems, each task is processed by respective processor in multicored computers e.g., household PCs which we can easily harness in recent years. If there is one policy to automatically decide the “best” combination and the number of processors (and computers), we effectively utilize those computational resources, thereby large number of jobs can be executed in parallel. In this paper, we propose a method for mapping of execution units for such environments. The method adopts a remapping technology after processor-execution unit mapping[6] is finished. Experimental comparisons by a simulation show the advantages of the proposed method.

QoS management method for W-LAN network considering hidden node issue

This paper proposes a new QoS management method for W-LAN network considering hidden node issue which handles the communication priority in accordance with each nodes required bit rate. By the proposed method a W-LAN node requiring the higher bit rates can have more chance to send data to control QoS requirement with propriety.