A New Multi-Resource Allocation Mechanism： A Tradeoff between Fairness and Efficiency in Cloud Computing

This paper addresses multi-resource fair allocation: a fundamental research topic in cloud computing. To improve resource utilization under well-studied fairness constraints, we propose a new allocation mechanism called Dominant Resource with Bottlenecked Fairness(DRBF), which generalizes Bottleneck-aware Allocation(BAA) to the settings of Dominant Resource Fairness(DRF). We classify users into different queues by their dominant resources. The goals are to ensure that users in the same queue receive allocations in proportion to their fair shares while users in different queues receive allocations that maximize resource utilization subject to well-studied fairness properties such as those in DRF. Under DRBF, no user 1) is worse off sharing resources than dividing resources equally among all users; 2) prefers the allocation of another user; 3) can improve their own allocation without reducing other users' allocations; and(4) can benefit by misreporting their resource demands. Experiments demonstrate that the proposed allocation policy performs better in terms of high resource utilization than does DRF.

Weighted De-Synchronization Based Resource Allocation in Wireless Networks

Considering the exponential growth of wireless devices with datastarving applications fused with artificial intelligence,the significance of wireless network scalability using distributed behavior and fairness among users is a crucial feature in guaranteeing reliable service to numerous users in the network environment.TheKuramoto model is described as nonlinear selfsustained phase oscillators spinning at varying intrinsic frequencies connected through the sine of their phase differences and displays a phase transition at a specific coupling strength,in which a mutual behavior is accomplished.In this work,we apply the Kuramoto model to achieve a weighted fair resource allocation in a wireless network,where each user has different quality of service(QoS)requirements.Because the original Kuramoto model is the synchronization model,we propose a new weighting parameter for representing requirement of each node resource and modify the Kuramoto model to achieveweighted fair resource allocation for users with different QoS requirements.The proposed modified Kuramoto model allocates all users the resource based on their weight among contending nodes in a distributed manner.We analyze the convergence condition for the proposed model,and the results reveal that the proposed algorithm achieves aweighted fair resource allocation and with potentially high convergence speed compared to previous algorithm.

RED-DTB: A Dual Token Bucket Based Queue Management Algorithm

Improving the Quality of Service (QoS) of Internet traffic is widely recognized as a critical issue for the next-generation networks. In this paper, we present a new algorithm for the active queue management, namely RED-DTB. This buffer control technique is used to enforce approximate fairness among a large number of concurrent Internet flows. Like RED (Random Early Detection) algorithm, the RED-DTB mechanism can be deployed to actively respond to the gateway congestion, keep the gateway in a healthy state, and protect the fragile flows from being stolen bandwidth by greedy ones. The algorithm is based on the so-called Dual Token Bucket (DTB) pattern. That is, on the one hand, every flow is rate-limited by its own token bucket, to ensure that it can not consume more than its fair share of bandwidth; On the other hand, to make some compensations to less aggressive flows, such as connections with larger round trip time or smaller sending window, and to gain a relatively higher system utilization coefficient, all flows, depending on their individual behavior, may have a chance to fetch tokens from the public token bucket when they run out of their own share of tokens. The algorithm is analyzed and evaluated by simulations, and is proved to be effective in protecting the gateway buffer and controlling the fair allocation of bandwidth among flows.

QoS-aware dynamic bandwidth allocation algorithm for Gigabit-capable PONS

The Gigabit-capable passive optical network(GPON)technology is being considered as a promising solution for the next-generation broadband access network.Since the network topology of the GPON is point-to-multipoint,a media access control called dynamic bandwidth allocation(DBA)algorithm is an important factor for determining the performance of the GPON.In this paper,we propose a new DBA algorithm to effectively and fairly allocate bandwidths among end users.This DBA algorithm supports differentiated services-a crucial requirement for a converged broadband access network with heterogeneous traffic.In this article we first reviewed the signaling and configuration of the DBA,and then proposed a new DBA scheme that implemented QoS-based priority for this need to maximally satisfy the requirements of all optical network units(ONUs)and provide differentiated services.Analyses and simulation results show that the new algorithm can improve the bandwidth utilization and realize the fairness for both different ONUs and services.

Improved resource allocation strategy in SU-CoMP network

Coordinated multi-point transmission and reception （CoMP） for single user, named as SU-CoMP, is considered as an efficient approach to mitigate inter-cell interference in orthogonal frequency division multiple access （OFDMA） systems. Two prevalent approaches in SU-CoMP are coordinated scheduling （CS） and joint processing （JP）. Although JP in SU-CoMP has been proved to achieve a great link performance improvement for the cell-edge user, efficient resource allocation （RA） on the system level is quite needed. However, so far limited work has been done considering JP, and most existing schemes achieved the improvement of cell-edge performance at cost of the cell-average performance degradation compared to the single cell RA. In this paper, a two-phase strategy is proposed for SU-CoMP networks. CS and JP are combined to improve both cell-edge and cell-average performance. Compared to the single cell RA, simulation results demonstrate that, the proposed strategy leads to both higher cell-average and cell-edge throughput.