Efficient Resource Allocation Mechanism Research Articles

Resource allocation mechanism for efficiency improvement in WDM networks: An analytical comparison study

This paper focuses on the performance improvement criteria suitable for WDM networks. Especially, an efficient resource allocation mechanism is introduced which organizes the number of data wavelengths into sets. Additionally, an effective access algorithm is applied by all stations in a decentralized way for the transmission over the data wavelengths sets, taking into account the receiver collisions phenomenon. The proposed wavelengths grouping configuration, in conjunction with the access algorithm, achieve more than 8% throughput increase and more than 10% delay decrease as compared to the single wavelengths set case, without requiring any additional synchronization or processing overhead. The performance measures of average throughput and average delay are analytically derived by means of closed mathematical formulas. The performance is extensively studied for several values of the performance parameters, while it is compared to the single wavelengths set case. Finally, the appropriate number of data wavelengths sets is determined in order for the throughput improvement to be optimum. The study results could be exploited to any multi-channel optical network configuration to improve the efficiency.

No user left behind: dynamic bottleneck-aware allocation of multiple resources

The fast-developing of cloud computing causes the resource management to the hot and heat research. Some researcher have studied the resource allocation and proposed some resource allocation mechanisms in the cloud computing, such as max–min fairness that is used in the data center. In order to satisfy the demand of cloud computing, we need to design a efficient and fair resource allocation mechanism. Wang et al. (Proceedings of the USENIX Conference on File and Storage Technologies (FAST), 229–242, 2014) proposed a new resource allocation mechanism, called balancing fairness and efficiency with bottleneck-aware allocation (BAA). BAA aims to find the fair between the users and maximize the resource utilization. However, BAA only consider the two resource types and the resource pool may have multiple resource types such as CPU, memory and storage. In addition, BAA consider the static allocation and do not take into account the dynamic allocation of users join the system one by one. To over this drawback, we propose the bottleneck-aware allocation of multiple resources (MRBAA) and dynamic bottleneck-aware allocation (DBBA) fair allocation mechanism. MRBAA and DBBA have lots of good properties. In addition, we characterizes the properties of our proposed mechanisms. Furthermore, our proposed mechanisms achieves the multiple resources fair and dynamic allocation to become more adaptable the real-world scenarios. Compared with the existing popular mechanism dominant resource fairness (DRF) from the literature, the simulation results show that our proposed mechanisms can efficient use of heterogeneous resources, increase multiple resources utilization, and schedule more tasks to benefit users.

A Comparative Analysis of Min-Min and Max-Min Algorithms based on the Makespan Parameter

In today’s computational environment, cloud computing has grown up as a utility. With increase in number of resource requests, there is a need for efficient resource allocation mechanism which aims at proper utilization of resources. Therefore in order to process these requests by making use of the available resources, an efficient task scheduling algorithm is required. An efficient task scheduling mechanism should be able to minimize completion time, maximize resource utilization and minimize makespan etc. Based on the efficiency metrics or parameters there are numerous type of existing scheduling algorithms but there is no such algorithm so far which can optimize all these parameters simultaneously, as each scheduling algorithm works on some particular parameter and tries to optimize it only. In this paper, two heuristic based algorithms: Min-Min and Max-Min have been described, simulated on CloudSim and then comparison based analysis has been done by taking makespan parameter into consideration. Keywords: Cloud Computing, Heuristic Algorithm, Max-Min Algorithm, Metatask, Min-Min Algorithm, Task Scheduling.

SDN-Enabled Game-Aware Routing for Cloud Gaming Datacenter Network

Cloud gaming or gaming as a service, the newest entry in the online gaming world, leverages the well-known concept of cloud computing to provide real-time gaming services to players. This gaming paradigm provides affordable, flexible, and high performance solutions for end users with constrained computing resources and enables them to play high-end graphic games on low-end thin clients, because it renders everything in the cloud and simply streams the resulting high-quality video to the player. Despite its advantages, cloud gaming’s quality of experience suffers from high and unstable end-to-end delay. According to this fact of cloud gaming, datacenters are in charge of performing complex rendering and video encoding computations, delivering high-quality gaming experience to gamers which requires an efficient and smart resource allocation mechanism to allot resources (e.g., memory and network bandwidth) to gaming sessions consistent with their requirements. In this paper, we propose a bi-objective optimization method to find an optimum path for packet transmission within a data center by minimizing delay and maximizing bandwidth utilization. We use a metaheuristic model, called analytic hierarchy process, to solve the NP-complete optimization problem. The resulting method is an analytic hierarchy process-based game aware routing (AGAR) scheme that considers requested game type and requirements in terms of delay and bandwidth to select the best routing path for a game session in a cloud gaming network. The method executes within a software defined network controller, which affords it a global view of the data center with respect to communication delay and available bandwidth. Simulation results indicate that the AGAR can reduce the end-to-end delay by up to 9.5% compared with three other conventional representative methods: the delay-based Dijkstra, the equal cost multi-path routing, and the Hedera routing algorithms. In addition, we demonstrate that the proposed method assigns game flows to network paths and OpenFlow switches in a balanced manner that prevents potential network bottlenecks.

Integration of Spatial Reuse and Allocation for Downlink Broadcast in LTE-Advanced and WiMAX Relay Networks

Downlink broadcast in LTE-Advanced-(LTE-A) and WiMAX-based relay networks is a crucial service for multimedia delivery. Currently, most research effort goes into devising efficient resource allocation mechanisms to achieve more efficient resource utilization. However, spatial reuse, which is an important technique in improving transmission capacity, has received little attention in the literature. Thus, in this work, we investigate ways of achieving more efficient resource allocation in wireless relay networks via spatial reuse. We first formulate a joint spatial reuse and resource allocation problem as an integer linear programming (ILP) model. We then consider a grouping mechanism in which relay stations (RSs) are grouped together if they do not interfere with each other's transmission signal. RSs in the same group can thus utilize spatial reuse by using the same set of resources to broadcast data. Because of the high computational complexity of the ILP model, we propose a two-phase heuristic solution. In the first phase, the enhanced-resource diminishing principle approach is employed to determine the number of resources required by the base station and a set of selected RSs. In the second phase, a max-coloring algorithm is employed to organize the selected RSs into broadcast groups and then assign the required resources to each group by exploring the maximum advantage of spatial reuse. The simulation results show that the proposed solution improves system performance up to 61%, as compared with two existing mechanisms.

Efficient Resource Allocation Mechanism for Federated Clouds

This study proposes a novel efficient resource allocation mechanism for federated clouds, which takes the communication overhead into consideration, to improve system throughput and reduce resource repacking overhead in the auto-scaling mechanism. In general, when the amount of service requests increases, more and more resources are allocated to satisfy these requests. However, single cloud cannot provide unlimited services with limited physical resources; therefore, the federation of multiple clouds may be one possible solution. In the federated cloud environment, when the workload changes, the resource allocation mechanism could adopt vertical/horizontal scaling fashions to repack the required resource into virtual machines. In the vertical scaling approach, the resource allocation mechanism allocates more resources into virtual machines for improving virtual machine's capability. In the horizontal scaling approach, the resource allocation mechanism allocates more virtual machines for enhancing the virtual cluster's capability. However, frequent resource repacking may reduce the system performance. Therefore, in order to improve system throughput and reduce repacking overhead, the proposed mechanism captures the execution pattern of applications by the profiling system and the resource status by the monitoring system, and then allocates resources for configuring the virtual cluster. Performance for NAS Parallel Benchmarks is evaluated. Experimental results show that the authors' approach could reduce repacking overhead and improve system throughput by comparing two previous works.

A Novel Multi-Dimension Resource Recycling Mechanism for Cloud Data Centers

Cloud servers can be started with ineffective resource arrangement, and extra costs are produced if unnecessary servers are started. This is a substantial fee for the cloud service provider. Therefore, each cloud data center needs an efficient resource allocation mechanism to prevent unnecessary cloud servers from being started. In general, resource allocation problems can be classified into either one-dimension or multi-dimension aspects. In 2013, the authors have proposed a multi-dimension resource allocation algorithm that can improve the utilization of cloud servers by as much as 97%. However, most of previous studies are more concerned with solving the resource allocation problem. In fact, after a period of time, the applications will finish their jobs, and resources been occupied on the servers will be released, thus decreasing the utilization of cloud servers. Therefore, this paper proposes a novel multi-dimension resource recycling algorithm to optimize the server resource again, to recycle the excess cloud resources and to reduce the unnecessary operating cloud servers.

An Efficient Resource Allocation Framework for Cloud Federations

With the evolution of Internet technology, cloud computing techniques are being used for many applications in our daily life. In particular, most large enterprises intend to help business groups reduce management cost and achieve higher benefits via cloud computing technology. In order to provide uninterrupted services to the clients and reduce the monetary cost of cloud services, the issue of retrieving borrowed resources efficiently has become increasingly important. In this study, we focus on the computation resource allocation in the cloud based environment. An efficient resource allocation mechanism conforming to cross-Identity Provider (IdP) architecture is proposed. We introduce an inter-trust relationship among clouds to allow a cloud to borrow available resources from external clouds via a constructed cloud federation. Solid simulation experiments are demonstrated to show the practicability and feasibility of our proposed mechanism in the cloud federation environment.DOI: http://dx.doi.org/10.5755/j01.itc.44.1.6875

Temporal-aware rate allocation in mission-oriented WSNs with sum-rate demand guarantee

Conventional resource allocation schemes for single-purpose WSNs are unable to capture the dynamicity of network and/or application specifications. This compels us to design mechanisms that support multi-purpose mission-oriented WSNs, where several missions with different characteristics coexist in an integrated WSN. The paper aims to propose an efficient temporal-aware resource allocation mechanism for mission-oriented WSNs. Furthermore, our design supports minimum sum-rate demand guarantee for active sessions of each mission. We use contention graph notion to analytically model the interference in our underlying WSN. Moreover, in-network processing operators play important role in our design scheme. Using a time-variant mission-oriented WSN model, we formulate our resource allocation problem as a dynamic network utility maximization problem, which aims to maximize a hybrid mix of an aggregated objective function of missions’ utilities and energy consumption, considering capacity, interference and sum-rate demand constraints. Finally, using dual decomposition method, we solve the problem and propose a distributed temporal-aware algorithm called TARC-SD that optimally allocates the shared bandwidth to the sources. The conducted experiments indicate that the TARC-SD algorithm achieves higher network utilization and wider feasible scenarios compared to conventional time-invariant schemes.

Optimization-based resource allocation in communication networks

The continuously growing number of applications competing for resources in current communication networks highlights the necessity for efficient resource allocation mechanisms to maximize user satisfaction. Optimization Theory can provide the necessary tools to develop such mechanisms that will allocate network resources optimally and fairly among users. The aim of this paper is to provide a starting point for researchers interested in applying optimization techniques in the resource allocation problem for current communication networks. To achieve that we, first, describe the fundamental optimization theory tools necessary to design optimal resource allocation algorithms. Then, we describe the Network Utility Maximization (NUM) framework, a framework that has already found numerous applications in network optimization, along with some recent advancements of the initial NUM framework. Finally, we summarize some of our recent work in the area and discuss some of the remaining research challenges towards the development of a complete optimization-based resource allocation protocol.

Dynamic resource allocation in fuzzy coalitions: a game theoretic model

We introduce an efficient and dynamic resource allocation mechanism within the framework of a cooperative game with fuzzy coalitions (cooperative fuzzy game). A fuzzy coalition in a resource allocation problem can be so defined that membership grades of the players in it are proportional to the fractions of their total resources. We call any distribution of the resources possessed by the players, among a prescribed number of coalitions, a fuzzy coalition structure and every membership grade (equivalently fraction of the total resources), a resource investment. It is shown that this resource investment is influenced by the satisfaction of the players in regard to better performance under a cooperative setup. Our model is based on the real life situations, where possibly one or more players compromise on their resource investments in order to help forming coalitions.

An Efficient Resource Allocation Mechanism for LTE–GEPON Converged Networks

Optical–wireless convergence is becoming popular as one of the most efficient access network designs that provides quality of service (QoS) guaranteed, uninterrupted, and ubiquitous access to end users. The integration of passive optical networks (PONs) with next-generation wireless access networks is not only a promising integration option but also a cost-effective way of backhauling the next generation wireless access networks. The QoS performance of the PON–wireless converged network can be improved by taking the advantages of the features in both network segments for bandwidth resources management. In this paper, we propose a novel resource allocation mechanism for long term evolution–Gigabit Ethernet PON (LTE–GEPON) converged networks that improves the QoS performance of the converged network. The proposed resource allocation mechanism takes the advantage of the ability to forecast near future packet arrivals in the converged networks. Moreover, it also strategically leverages the inherited features and the frame structures of both the LTE network and GEPON, to manage the available bandwidth resources more efficiently. Using extensive simulations, we show that our proposed resource allocation mechanism improves the delay and jitter performance in the converged network while guarantying the QoS for various next generation broadband services provisioned for both wireless and wired end users. Moreover, we also analyze the dependency between different parameters and the performance of our proposed resource allocations scheme.

Designing Efficient Resource Procurement and Allocation Mechanisms in Humanitarian Logistics

We analyze the efficacy of different asset transfer mechanisms and provide policy recommendations for the design of humanitarian supply chains. As a part of their preparedness effort, humanitarian organizations often make decisions on resource investments ex-ante because doing so allows for rapid response if an adverse event occurs. However, programs typically operate under funding constraints and donor earmarks with autonomous decision-making authority resting with the local entities, which makes the design of efficient humanitarian supply chains a challenging problem. We formulate this problem in an agency setting with two independent aid programs, where different asset transfer mechanisms are considered and where investments in resources are of two types: a primary resource that is needed for providing the aid, and infrastructural investments that improve the operation of the aid program in using the primary resource. The primary resource is modeled as either a divisible or indivisible good, and is acquired from earmarked donations. We show that allowing aid programs the exibility of transferring primary resources improves the efficiency of the system by yielding greater social welfare than when this exibility does not exist. More importantly, we show that a central entity that can acquire primary resources from one program and sell them to the other program can further improve system efficiency by providing a mechanism that facilitates the transfer of primary resources and eliminates losses from gaming. This outcome is achieved without depriving the individual aid programs of their decision-making autonomy while maintaining the constraints under which they operate. We find that outcomes with centralized resource transfer but decentralized infrastructural investments by the aid programs are the same as with a completely centralized system (where both resource transfer and infrastructural investments are centralized).

Efficient and incentive-compatible resource allocation mechanism for P2P-assisted content delivery systems

In this paper, we propose an efficient resource allocation mechanism for Peer-to-peer(P2P) assisted content delivery systems. The key component of the mechanism is a novel distributed data bartering protocol that can redistribute the bandwidth resources among peers according to their contributions. It is formally proved that the proposed mechanism is incentive-compatible, i.e., the only optimal strategy of each peer is to behave honestly and to contribute as much as possible. Based on the proposed mechanism, we design a general P2P-assisted content delivery system for delivering both live video streams and static files. We also propose the design of some key modules including peer/data organization, bandwidth allocation, data scheduling, and tax rate adjustment. We implement our system on a packet-level simulator and simulate two scenarios: P2P live streaming and P2P file sharing. Experimental results on real world traces show that the system can provide both differentiated services and satisfactory performance to peers. Furthermore, when bandwidth resource levels are dynamic, the system can still guarantee consistent service quality by adjusting the global tax rate adaptively. We also find that our system can effectively defend against dishonest behaviors even when the percentage of cheating peers is as high as 50%.

Negotiation strategies considering market, time and behavior functions for resource allocation in computational grid

Providing an efficient resource allocation mechanism is a challenge to computational grid due to large-scale resource sharing and the fact that Grid Resource Owners (GROs) and Grid Resource Consumers (GRCs) may have different goals, policies, and preferences. In a real world market, various economic models exist for setting the price of grid resources, based on supply-and-demand and their value to the consumers. In this paper, we discuss the use of multiagent-based negotiation model for interaction between GROs and GRCs. For realizing this approach, we designed the Market- and Behavior-driven Negotiation Agents (MBDNAs). Negotiation strategies that adopt MBDNAs take into account the following factors: Competition, Opportunity, Deadline and Negotiator's Trading Partner's Previous Concession Behavior. In our experiments, we compare MBDNAs with MDAs (Market-Driven Agent), NDF (Negotiation Decision Function) and Kasbah in terms of the following metrics: total tasks complementation and budget spent. The results show that by taking the proposed negotiation model into account, MBDNAs outperform MDAs, NDF and Kasbah.

Grid Commerce, Market-Driven G-Negotiation, and Grid Resource Management

Although the management of resources is essential for realizing a computational grid, providing an efficient resource allocation mechanism is a complex undertaking. Since Grid providers and consumers may be independent bodies, negotiation among them is necessary. The contribution of this paper is showing that market-driven agents (MDAs) are appropriate tools for Grid resource negotiation. MDAs are e-negotiation agents designed with the flexibility of: 1) making adjustable amounts of concession taking into account market rivalry, outside options, and time preferences and 2) relaxing bargaining terms in the face of intense pressure. A heterogeneous testbed consisting of several types of e-negotiation agents to simulate a Grid computing environment was developed. It compares the performance of MDAs against other e-negotiation agents (e.g., Kasbah) in a Grid-commerce environment. Empirical results show that MDAs generally achieve: 1) higher budget efficiencies in many market situations than other e-negotiation agents in the testbed and 2) higher success rates in acquiring Grid resources under high Grid loadings.

Modeling and generating realistic streaming media server workloads

Currently, Internet hosting centers and content distribution networks leverage statistical multiplexing to meet the performance requirements of a number of competing hosted network services. Developing efficient resource allocation mechanisms for such services requires an understanding of both the short-term and long-term behavior of client access patterns to these competing services. At the same time, streaming media services are becoming increasingly popular, presenting new challenges for designers of shared hosting services. These new challenges result from fundamentally new characteristics of streaming media relative to traditional web objects, principally different client access patterns and significantly larger computational and bandwidth overhead associated with a streaming request. To understand the characteristics of these new workloads we use two long-term traces of streaming media services to develop MediSyn, a publicly available streaming media workload generator . In summary, this paper makes the following contributions: (i) we propose a framework for modeling long-term behavior of network services by capturing the process of file introduction, non-stationary popularity of media accesses, file duration, encoding bit rate, and session duration. (ii) We propose a variety of practical models based on the study of the two workloads. (iii) We develop an open-source synthetic streaming service workload generator to demonstrate the capability of our framework to capture the models.

A note on some properties of an efficient network resource allocation mechanism

We present some limiting properties of a network resource allocation mechanism known as the Progressive Second Price (PSP) auction. This mechanism aims at efficiently allocate network resources, such as bandwidth or buffer capacity, in an environment characterized by competing users; the PSP auction seeks to solve or at least to ameliorate congestion in a network demanding a low signalling burden between the auctioneer and the users, and solving the allocation problem of an (theoretically) infinitely divisible resource. The allocation rule is inspired in the second price (Vickrey) auction. Our analysis of the PSP auction explores its limiting properties, namely, how the allocation changes in the presence of a polarized set of users. A polarized set of users is a mixture of users of two types: high valuation, low demand users and low valuation, high demand users. Mechanisms such as auctions are becoming increasingly popular to handle the resource allocation problem in networks facing congestion, such as the access to Internet-based services.

A note on dependable real-time communication in multihop networks

The issue of providing fault-tolerance in real-time communication has been a problem of growing importance. There are two basic approaches for satisfying fault-tolerant requirements in real-time communication: (i) forward error recovery approach and (ii) detect and recovery approach. The first approach is well-suited for hard real-time communication, whereas the second approach is well-suited for soft real-time communication. Neither of these basic approaches is well-suited for supporting both hard and soft real-time communication. In this paper, we propose an integrated scheme that not only supports such a mixed communication requirements, but also improves the call acceptance rate significantly due to its efficient resource allocation mechanisms such as traffic dispersion and backup multiplexing. The effectiveness of the proposed scheme has been evaluated through extensive simulation studies.

Continuous-media courseware server: a study of client interactions

Considerable research has gone into investigating networking and operating system mechanisms to support the transfer and playout of stored continuous media, but there is very little information available about how users actually interact with such systems. Developing a user workload characterization can help in the design and evaluation of efficient CM resource allocation and access mechanisms. The authors developed an interactive Web-based, multimedia, client-server application, known as the Multimedia Asynchronous Networked Individualized Courseware, or MANIC, which streams synchronized CM (currently audio) and HTML documents to remote users. This article presents empirical and analytic characterizations of observed user session-level behavior (for example, the length of individual sessions) and interactive behavior (for example, the time between starting, stopping, and pausing the audio within a session). The data come from a full-semester senior-level course given by the University of Massachusetts to more than 200 students who used MANIC to listen to the stored audio lectures and to view the lecture notes.