Optimal load distribution and speed scaling in a heterogeneous data center that takes account of the tradeoff between the maintenance and operation costs and system performance are crucial issues of cloud computing. Due to the changing states of cloud centers and diversity of the task arrival rate, a static control model is infeasible for cloud computing. In this article, we aim to provide a novel multi-server control model with dynamic feedback to acquire dynamic states of the cloud system, and with queue waiting cost-awareness to optimize the queue wait time and load distribution in task assignment and server configuration management. Using the technology of speed scaling, each server in a data center is configured as an$M/M/1$queue system with variable service rate, and the service rate is a function of the length of the task queue. We formulate two optimization problems, the optimal load distribution problem and the optimal service rate controlling problem, and provide algorithms to solve these problems, facilitating load distributions and service rate adjustments. We also present numerical simulations to validate our model. The results show our model to be efficient in multi-server dynamic configurations and task assignments according to the feedback information for the tradeoff between the system cost and performance.

Proactive preparedness to cope with extreme weather events is significantly helpful in reducing the restoration cost and enhancing the resilience of distribution systems. This paper is focused on the resource allocation problem in distribution systems ahead of a coming hurricane. Generation resources such as diesel oil and batteries are considered for allocation, which can be used to serve outage critical load in the post-hurricane restoration. Electric buses are also considered as a kind of resource. Considering the uncertainties of system faults, the allocation problem is formulated into a mixed-integer stochastic nonlinear program. A heuristic method is then proposed, which obtains the allocation plan by solving a mixed-integer linear program. Numerical simulations are performed on the IEEE 123-node feeder system under several scenarios to demonstrate the effectiveness of the proposed method. The impacts of resources transportation cost, initial distribution of electric buses, and hurricane severity on the allocation plan are discussed.

In our previous studies, an optical-code-division multiplexing (OCDM)-based network architecture has been proposed to improve the wavelength utilization and to provide finer bandwidth granularities to users. However, since up-to-date all-optical regeneration technologies are still immature, signal quality degradation at intermediate nodes is considered as a crucial problem in transparent optical networks. Because the remaining noises may propagate, and be accumulated along the light-path, which eventually result in unacceptable signal quality. In this paper, we first define a cycle attack problem due to multi-access interference (MAI) propagation in transparent OCDM-based networks, which may cause severe service disruption. A depth-first-search (DFS)-based algorithm is proposed to diagnose such cycle attacks and a heuristic wavelength assignment called advanced first-fit (AFF) scheme taking cycle attacks into account is found to be a simple but effective approach mitigating the cycle attack impacts by computer simulations.

In this paper, an OCDM-based multi- granularity optical cross-connect (MG-OXC) is proposed to enable the optical path switching in bandwidth granularity finer than the wavelength-based path. Such an optical path switched by OCDM-based MG-OXC is called OCDM label-switched path (OCDM-LSP), which is capable of hybrid optical code (OC)/wavelength conversion along the path from a source to destination. With distinct OCs, OCDM-LSPs can share a single wavelength without any synchronization mechanism and be discriminated by ultra- fast optical correlation. The asynchronous ability and OC conversion distinguish the OCDM-LSP from Packet-LSP or TDM-LSP in resolving granularity problem in wavelength- based path. Based upon our proposal, all-optical optical path provisioning including F-LSP, WB-LSP, L-LSP and OCDM- LSP can be realized in the transport layer. Simulation is conducted and the results show the proposed OCDM-switching has gained a 2 order of blocking probability reduction even though only 5 OCs are employed. Furthermore, OC conversion in OCDM-LSP is proved to be another benefit for high flexibility in optical path provisioning and blocking performance improvement comparing to TDM-LSP.