Yahaya Coulibaly

Resource scheduling for infrastructure as a service (IaaS) in cloud computing

Resource scheduling assigns the precise and accurate task to CPU, network, and storage. The aim behind this is the extreme usage of resources. However, well organized scheduling is needed for both cloud providers and cloud users. This paper is a chronological study of recent issues of resource scheduling in IaaS cloud computing environment. In our study, we investigate resource scheduling schemes and algorithms used by different researchers and categorize these approaches on the basis of problems addressed, schemes and the parameters used in evaluating different approaches. Based on various studies considered in this survey, we perceive that different schemes and algorithms did not consider some essential parameters and enhancement is requisite to improve the performance of the existing schemes. Furthermore, this study will trigger new and innovative methods of handling the problems of resource scheduling in the cloud and will help researchers in understanding the existing methodologies for the future adaptation and enhancements. Presents a review of resource scheduling techniques in IaaS cloud.Classify techniques on the basis of problems addressed and schemes used.Analyzes the parameters used to evaluate techniques used in several studies.Suggests future research areas in cloud computing, regards to resource scheduling.

Recent advancements in resource allocation techniques for cloud computing environment: a systematic review

There are two actors in cloud computing environment cloud providers and cloud users. On one hand cloud providers hold enormous computing resources in the cloud large data centers that rent the resources out to the cloud users on a pay-per-use basis to maximize the profit by achieving high resource utilization. On the other hand cloud users who have applications with loads variation and lease the resources from the providers they run their applications within minimum expenses. One of the most critical issues of cloud computing is resource management in infrastructure as a service (IaaS). Resource management related problems include resource allocation, resource adaptation, resource brokering, resource discovery, resource mapping, resource modeling, resource provisioning and resource scheduling. In this review we investigated resource allocation schemes and algorithms used by different researchers and categorized these approaches according to the problems addressed schemes and the parameters used in evaluating different approaches. Based on different studies considered, it is observed that different schemes did not consider some important parameters and enhancement is required to improve the performance of the existing schemes. This review contributes to the existing body of research and will help the researchers to gain more insight into resource allocation techniques for IaaS in cloud computing in the future.

EDR: efficient data routing in wireless sensor networks

Routing in sensor networks is very challenging compared to contemporary communication and wireless ad-hoc networks. This is attributed to several characteristics that distinguish them from such networks. Transmitting data from every sensor node within the deployment region might result in significant and unnecessary redundancy in data and incur avoidable energy and traffic expenditure; thus, a routing protocol which is able to select a set of sensor nodes and utilise data aggregation during the relaying of data is of great importance. Furthermore, timely delivery of data is required in many applications of Wireless Sensor Networks (WSNs), such as real-time target tracking in battle environments, emergent event triggering in monitoring applications impose the need of efficient routing protocol. In this paper, we propose a data routing scheme which ensures increased network lifetime by exploiting energy efficiently, supporting reliability, fast delivery of delay-sensitive data and achieving low-cost sensor design. Simulation results demonstrate that, the proposed routing scheme outperforms related existing approaches in terms of all the evaluation metrics such as packet delivery ratio, energy efficiency and timely delivery of data.

QOS-aware error recovery in wireless body sensor networks using adaptive network coding.

Wireless body sensor networks (WBSNs) for healthcare and medical applications are real-time and life-critical infrastructures, which require a strict guarantee of quality of service (QoS), in terms of latency, error rate and reliability. Considering the criticality of healthcare and medical applications, WBSNs need to fulfill users/applications and the corresponding networks QoS requirements. For instance, for a real-time application to support on-time data delivery, a WBSN needs to guarantee a constrained delay at the network level. A network coding-based error recovery mechanism is an emerging mechanism that can be used in these systems to support QoS at very low energy, memory and hardware cost. However, in dynamic network environments and user requirements, the original non-adaptive version of network coding fails to support some of the network and user QoS requirements. This work explores the QoS requirements of WBSNs in both perspectives of QoS. Based on these requirements, this paper proposes an adaptive network coding-based, QoS-aware error recovery mechanism for WBSNs. It utilizes network-level and user-/application-level information to make it adaptive in both contexts. Thus, it provides improved QoS support adaptively in terms of reliability, energy efficiency and delay. Simulation results show the potential of the proposed mechanism in terms of adaptability, reliability, real-time data delivery and network lifetime compared to its counterparts.

Quantifying the severity of blackhole attack in wireless mobile Adhoc networks

Blackhole attack is one of the severe attacks in MANET routing protocols. Generation of this attack is simple, which does not require specific tools or sophisticated attack techniques. However, this attack seriously corrupts the routing tables of nodes in the network. Even worse, the attack could increase the chances of losing confidential data and deny network services. Many researchers have proposed a variety of solutions to prevent this Conventional Blackhole Attack (CBA). Unfortunately, none of them has measured the severity of this attack. Filling gaps in measuring the severity of this attack, this research proposes a new security metrics, namely Corruption Routing Table (CRT), Compromising Relay Node (CRN) and Compromising Originator Node (CON). In addition, this research also introduces a new blackhole attack, namely Hybrid Black Hole Attack (HBHA) with two variants — Independent and Cooperative HBHAs. The proposed metrics proved effective in measuring the severity of both CBA and HBHAs. Simulation using Java in Time Simulator/Scalable Wireless Adhoc Network Simulator (JiST/SWANS) showed that Independent HBHA is the most severe attack compared to Cooperative HBHA and CBA. In addition, Cooperative HBHA is the most efficient attack than Independent HBHA and CBA.

A review of blackhole attack in mobile adhoc network

Mobile ad hoc network (MANET) is used in a wide range of application from civilian usage to military battlefield communications. Using multi-hop communication model, a source node can directly communicate with the destination node without infrastructure. Unfortunately, this feature causes the network vulnerable to various attacks. Security preventions implemented in the traditional wired network are unsuitable for MANET. This paper studies several types of attacks, and then discusses one of the most severe attack in the network, blackhole attack. Furthermore, recent efforts for preventing the blackhole attack are comprehensively discussed. Study of literature indicates that the prevention of the blackhole attack can be categorized into three groups, i.e., Protection based on Cryptography, Protocol Modification, and Intrusion Detection and Counter Measure.

Distance based Reliable and Energy Efficient (DREE) Routing Protocol for Underwater Acoustic Sensor Networks

The distinctive features of Underwater Acoustic Sensor Networks (UW-ASN) such as high propagation delay, asymmetric channels, high errors rates and limited bandwidth cause significant issues due to the use of acoustic medium as a method of communication. Therefore, designing an efficient underwater protocol that make efficient use of available battery resources and provides reliability in unreliable acoustic environment is a challenging task. In this paper, we therefore present a distance based reliable and energy efficient routing protocol (DREE). The contribution in this routing protocol is twofold 1) DREE actually takes in to account physical distance unlike similar protocol R-ERP 2 R where physical distance has not been utilized 2) DREE utilizes Fuzzy logic based link Quality Estimator (F-LQE) which has not been put on the test before in underwater networks. Moreover, DREE extends network lifetime not only by providing energy balancing but by reducing number of transmissions required to deliver a packet successfully. Experiments performed in NS2 simulator prove that DREE contributes to improve results in comparison with R-ERP 2 R which is the similar and only distance based reliable protocol present in literature and well known depth based routing protocol DBR

Fuzzy-Based Adaptive Hybrid Burst Assembly Technique for Optical Burst Switched Networks

The optical burst switching (OBS) paradigm is perceived as an intermediate switching technology for future all-optical networks. Burst assembly that is the first process in OBS is the focus of this paper. In this paper, an intelligent hybrid burst assembly algorithm that is based on fuzzy logic is proposed. The new algorithm is evaluated against the traditional hybrid burst assembly algorithm and the fuzzy adaptive threshold (FAT) burst assembly algorithm via simulation. Simulation results show that the proposed algorithm outperforms the hybrid and the FAT algorithms in terms of burst end-to-end delay, packet end-to-end delay, and packet loss ratio.

A Novel Routing Optimization in Optical Burst Switching Networks

Among the three optical network technologies being studied for future Internet, namely, Optical Circuit Switching, Optical Packet switching and Optical Burst switching; Optical Burst Switching is, up to now, the most likely to be implemented in the near future. Nevertheless, bursts contention persists and more work need to be done to minimize bursts loss before Optical Burst switching can be implemented. In this paper, we propose efficient bandwidth utilization and routing optimization scheme aiming at avoiding bursts contention and making better use of the large bandwidth offered by fiber links through Wavelength Division Multiplexing and Dense Wavelength Division Multiplexing. The proposed solution is based on the Hierarchical Optical Burst switching techniques and a modified hybrid algorithm of Ant Colony Optimization and Genetic Algorithm, taking into account the Streamline Effect in Optical Burst switching. Through simulation, we expect low burst lost probability and efficient use of bandwidth.

A survey of grid-based searching techniques for large scale distributed data

The large-scale distributed dataset searching faced dynamicity, heterogeneity, and latency issues that emphasize the importance of approach to orchestrate the search operations. The growth of grid computing gives a reliable environment for the effective usage of this distributed huge dataset. In order to build effective and efficient searching technique, varieties of researches were developed based on grid technology. The goal of such researches is to solve interoperability and heterogeneous of the data sets, and increase the efficiency of search techniques by harnessing the grid computing capabilities. This paper reviews and analyzes the search techniques carried out in the domain of large-scale dataset search based on a grid. Several searching approaches are studied to identify which attributes are ignored and which are considered. The techniques are evaluated depending on the aforementioned attributes and are presented in a table. The contribution of this study is to illustrate the characteristics, capabilities and limitations of existing solutions that can guide the researchers in this evolving area. HighlightsIssues involved in Grid-Based massive data search techniques are being discussed.Criteria such as real time search, online data, and search technique are explained.Features such as research Field, UI, automate grid Tasks, architecture are discussed.Explanation table to Identify Open issues and help future data search techniques.