The container loading problem (CLP) is a well-known NP-hard problem. Due to the computation complexity, heuristics is an often-sought approach. This article proposes two heuristics to pack homogeneous rectangular boxe...The container loading problem (CLP) is a well-known NP-hard problem. Due to the computation complexity, heuristics is an often-sought approach. This article proposes two heuristics to pack homogeneous rectangular boxes into a single container. Both algorithms adopt the concept of building layers on one face of the container, but the first heuristic determines the layer face once for all, while the second treats the remaining container space as a reduced-sized container after one layer is loaded and, hence, selects the layer face dynamically. To handle the layout design problem at a layer's level, a block-based 2D packing procedure is also developed. Numerical studies demonstrate the efficiency of the heuristics.展开更多
Virtualization is a common technology for resource sharing in data center. To make efficient use of data center resources, the key challenge is to map customer demands （modeled as virtual data center, VDC） to the ph...Virtualization is a common technology for resource sharing in data center. To make efficient use of data center resources, the key challenge is to map customer demands （modeled as virtual data center, VDC） to the physical data center effectively. In this paper, we focus on this problem. Distinct with previous works, our study of VDC embedding problem is under the assumption that switch resource is the bottleneck of data center networks （DCNs）. To this end, we not only propose relative cost to evaluate embedding strategy, decouple embedding problem into VM placement with marginal resource assignment and virtual link mapping with decided source-destination based on the property of fat-tree, but also design the traffic aware embedding algorithm （TAE） and first fit virtual link mapping （FFLM） to map virtual data center requests to a physical data center. Simulation results show that TAE＋FFLM could increase acceptance rate and reduce network cost （about 49% in the case） at the same time. The traffie aware embedding algorithm reduces the load of core-link traffic and brings the optimization opportunity for data center network energy conservation.展开更多
Two new methods were presented for power flow tracing(PFT).These two methods were compared and the results were discussed in detail.Both methods use the active and reactive power balance equations at each bus in order...Two new methods were presented for power flow tracing(PFT).These two methods were compared and the results were discussed in detail.Both methods use the active and reactive power balance equations at each bus in order to solve the tracing problem.The first method considers the proportional sharing assumption while the second one uses the circuit laws to find the relationship between power inflows and outflows through each line,generator and load connected to each bus of the network.Both methods are able to handle loop flow and loss issues in tracing problem.A formulation is also proposed to find the share of each unit in provision of each load.These methods are applied to find the producer and consumer's shares on the cost of transmission for each line in different case studies.As the results of these studies show,both methods can effectively solve the PFT problem.展开更多
The resource allocation problem in data centre networks refers to a map of the workloads provided by the cloud users/tenants to the Substrate Network(SN)which are provided by the cloud providers.Existing studies consi...The resource allocation problem in data centre networks refers to a map of the workloads provided by the cloud users/tenants to the Substrate Network(SN)which are provided by the cloud providers.Existing studies consider the dynamic arrival and departure of the workloads,while the dynamics of the substrate are ignored.In this paper,we first propose the resource allocation with the dynamic SN,and denote it as GraphMap-DS.Then,we propose an efficient mapping algorithm for GraphMap-DS.The performance of the proposed algorithm is evaluated by performing simulation experiments.Our results show that the proposed algorithm can effectively solve the GraphMap-DS.展开更多
基金the Natural Sciences and Engineering Research Council of Canada (NSERC) under the Discovery Grant Program
文摘The container loading problem (CLP) is a well-known NP-hard problem. Due to the computation complexity, heuristics is an often-sought approach. This article proposes two heuristics to pack homogeneous rectangular boxes into a single container. Both algorithms adopt the concept of building layers on one face of the container, but the first heuristic determines the layer face once for all, while the second treats the remaining container space as a reduced-sized container after one layer is loaded and, hence, selects the layer face dynamically. To handle the layout design problem at a layer's level, a block-based 2D packing procedure is also developed. Numerical studies demonstrate the efficiency of the heuristics.
基金This research was partially supported by the National Grand Fundamental Research 973 Program of China under Grant （No. 2013CB329103）, Natural Science Foundation of China grant （No. 61271171）, the Fundamental Research Funds for the Central Universities （ZYGX2013J002, ZYGX2012J004, ZYGX2010J002, ZYGX2010J009）, Guangdong Science and Technology Project （2012B090500003, 2012B091000163, 2012556031）.
文摘Virtualization is a common technology for resource sharing in data center. To make efficient use of data center resources, the key challenge is to map customer demands （modeled as virtual data center, VDC） to the physical data center effectively. In this paper, we focus on this problem. Distinct with previous works, our study of VDC embedding problem is under the assumption that switch resource is the bottleneck of data center networks （DCNs）. To this end, we not only propose relative cost to evaluate embedding strategy, decouple embedding problem into VM placement with marginal resource assignment and virtual link mapping with decided source-destination based on the property of fat-tree, but also design the traffic aware embedding algorithm （TAE） and first fit virtual link mapping （FFLM） to map virtual data center requests to a physical data center. Simulation results show that TAE＋FFLM could increase acceptance rate and reduce network cost （about 49% in the case） at the same time. The traffie aware embedding algorithm reduces the load of core-link traffic and brings the optimization opportunity for data center network energy conservation.
文摘Two new methods were presented for power flow tracing(PFT).These two methods were compared and the results were discussed in detail.Both methods use the active and reactive power balance equations at each bus in order to solve the tracing problem.The first method considers the proportional sharing assumption while the second one uses the circuit laws to find the relationship between power inflows and outflows through each line,generator and load connected to each bus of the network.Both methods are able to handle loop flow and loss issues in tracing problem.A formulation is also proposed to find the share of each unit in provision of each load.These methods are applied to find the producer and consumer's shares on the cost of transmission for each line in different case studies.As the results of these studies show,both methods can effectively solve the PFT problem.
基金supported by the National Basic Research of China(973 Program)under Grant No.2011CB302601the National Natural Science Foundation of China under Grants No.90818028,No.6903043,No.61202117the National High Technology Research and Development Program of China(863 Program)under Grant No.2012AA011201
文摘The resource allocation problem in data centre networks refers to a map of the workloads provided by the cloud users/tenants to the Substrate Network(SN)which are provided by the cloud providers.Existing studies consider the dynamic arrival and departure of the workloads,while the dynamics of the substrate are ignored.In this paper,we first propose the resource allocation with the dynamic SN,and denote it as GraphMap-DS.Then,we propose an efficient mapping algorithm for GraphMap-DS.The performance of the proposed algorithm is evaluated by performing simulation experiments.Our results show that the proposed algorithm can effectively solve the GraphMap-DS.
