This paper addresses network slicing in a large-scale Multi-Access Edge Computing (MEC)-enabled Radio Access Network (RAN) comprising heterogeneous edge nodes with varying computing and storage resource capacities. These resources are dynamically allocated to slice requests and released when the service of a slice request is completed. Our objective is to optimize the resource allocation for each admitted arriving slice request, considering its demands for computing and storage resources, to maximize the long-run average Earning Before Interest and Taxes (EBIT) of the MEC slicing system. We formulate the optimization problem as a Restless Multi-Armed Bandit (RMAB)-based resource allocation problem with a nonlinear cost rate function. To solve this, we introduce a new policy called Prioritizing-the-Future-Approximated earning per request (PFA) where for each admitted slice request, we always prioritize the allocation of the resource combination that gives the highest achievable earning, considering the future effects of this allocation. PFA is designed to be scalable and applicable to large-scale networks. We numerically demonstrate the superior performance of PFA in maximizing long-run average EBIT through simulations, comparing it with two baseline policies, at various cases of parameter values. Moreover, our findings offer insights for network operators in resource allocation policy selection.

In design and optimization of intensive care unit (ICU) networks, one common practice is to prioritize the treatment for patients of higher emergency levels, while ensuring fairness to other patients by guaranteeing a certain Quality of Service (QoS) level. One common approach to realize such priority arrangement is bed reservation policy, which designates a certain number of last occupied beds in each hospital to be exclusively used by certain patient classes. In this paper, we propose an approach that can significantly improve the computational efficiency in obtaining the optimal reservation thresholds for each patient class, given their respective requirements, in a non-hierarchical ICU model (where the external emergency patients can possibly be allocated to any ICU hospital), which has been shown to be computationally challenging in performance evaluation and optimization. Specifically, we apply the Information Exchange Surrogate Approximation (IESA) to analytically approximate the key QoS metrics under given reservation thresholds, and the integer Particle Swarm Optimization (PSO) algorithm to search for the optimal threshold based on the approximation results by IESA. We demonstrate numerically, with real data from ICUs in Hong Kong, that IESA can obtain reasonably accurate results for QoS metrics, and thus lead to accurate optimal reservation thresholds. In addition, our proposed approach combining IESA and PSO can significantly reduce the computation time by more than four orders of magnitude compared to the state-of-the-art evaluation and optimization methods in existing research for similar problems, especially for ICU networks with practical sizes.

We focus on energy-efficient offloading strategies in a slicing-enabled large-scale edge network, or an "edge slicing" system, with different computing/storage components, on which the service capacities are dynamically released and reused by the incoming user requests. The offloading problem is challenged by its large problem size and the heterogeneity of the service components and user requests, leading to the high-dimensional state space of the underlying stochastic process. We formulate the problem in the manner of the restless-bandit-based (RB-based) resource allocation problem and generalize unrealistic previously made assumptions on specific forms of the power functions, such as linearity, convexity, and taking only binary power states. We adapt the RB-based resource allocation technique to the offloading problem. We quantify actions of selecting certain service components to serve requests through marginal rewards, which take consideration of both the history and future effects of the corresponding action. The marginal rewards exist in closed forms when assuming linear power functions, but it remains an open question in the non-linear case. We approximate the marginal rewards by introducing state-dependent coefficients that compensate for the undesirable effects of non-linearity. We propose a scheduling policy that always prioritizes the service components with the highest marginal rewards, which is simple and applicable in the large-scale case. In the special case with linear power functions, the policy becomes asymptotically optimal - it approaches optimality when the number of components tends to infinity. We numerically demonstrate the effectiveness and robustness of the proposed policy in practical situations with respect to energy efficiency.

Federated learning (FL) is an emerging technique aiming at improving communication efficiency in distributed networks, where many clients often request to transmit their calculated parameters to an FL server simultaneously. However, in wireless networks, the above mechanism may lead to prolonged transmission time due to unreliable wireless transmission and limited bandwidth. This paper proposes a communication scheme to minimize the uplink transmission time for FL in wireless networks. The proposed approach consists of two major elements, namely a modified Ring All-reduce (MRAR) architecture that integrates D2D wireless communications to facilitate the commu-nication process in FL, and applies an Ant Colony Optimization-based algorithm to identify the optimal composition of the MRAR architecture. Numerical results show that our proposed approach is robust and can significantly reduce the transmission time compared to the conventional star topology. Notably, the reduction in uplink transmission time compared to baseline policies can be substantial in scenarios applicable to large-scale FL, where client devices are densely distributed.

We consider a wireless network scenario applicable to metropolitan areas with developed public transport networks and high commute demands, where the mobile user equipments (UEs) move along fixed and predetermined trajectories and request to associate with millimeter-wave (mmWave) base stations (BSs). An effective and efficient algorithm, called the Sequence Q-learning Algorithm (SQA), is proposed to maximize the long-run average transmission rate of the network, which is an NP-hard problem. Furthermore, the SQA tackles the complexity issue by only allowing possible re-associations (handover of a UE from one BS to another) at a discrete set of decision epochs and has polynomial time complexity. This feature of the SQA also restricts too frequent handovers, which are considered highly undesirable in mmWave networks. Moreover, we demonstrate by extensive numerical results that the SQA can significantly outperform the benchmark algorithms proposed in existing research by taking all UEs’ future trajectories and possible decisions into account at every decision epoch.

We propose a versatile cross-layer framework to analyze performance metrics for mobile traffic in fifth-generation (5G) millimeter wave (mmWave) networks. Our proposed framework is based on stochastic geometry, teletraffic models, and the classical Erlang Fixed Point Approximation method, with the objective of evaluating blocking probability, mean service time of user requests, and utilization rate of base stations by taking into account practical concerns in mmWave networks including blockages encountered by mmWaves in the physical layer, capacity constraints in the network layer, and the stochastic nature of mobile traffic. We demonstrate by numerical results that our analytical method is accurate and computationally-efficient.

Wireless network slicing is a promising technology for next-generation networks to provide tailored on-demand services to mobile users. We consider a scheduling policy for wireless network slicing with the aim to maximize the energy efficiency of the network defined as the ratio of long-run average throughput of user requests to the long- run average power consumption. This gives rise to a problem of extremely high computational complexity which prevents direct application of conventional optimization techniques. We propose a scalable priority-based policy, referred to as the Most Energy-Efficient Resource First (MEERF). MEERF is proved to be asymptotically optimal in the special case appropriate for a local wireless environment with highly dense user population and exponentially distributed service time requirement. The robustness of MEERF to different service time distributions is demonstrated by extensive simulations. We present numerically the effectiveness of MEERF %balancing the QoS and relevant power consumption by comparing it with benchmark policies in a more general network with potentially geographically distributed users and infrastructures. The results show that MEERF outperforms the benchmark policies in most of our experiments and achieves up to 52% improvement in terms of energy efficiency.

With rapid development and high penetration of renewable energy, distributed generation, and energy storage system, as well as the emerging of electric vehicles and responsive loads, numerous uncertainties are brought into modern power systems. The power system robust operation become much more complicated than before. So, there is a urgent demand to develop a "faster than real time" power system Energy Management System (EMS) analysis tool using advanced data management and analysis technologies. In this paper, the power system EMS real time analysis framework based on the evolving graph is proposed. The power system is modeled as an evolving graph and its network analysis is implemented on a graph database platform. Testing results with a provincial power system demonstrate that the proposed analysis framework can significantly improve the performance and reach the goal of real-time power system analysis.

Energy efficiency and Quality of Service (QoS) are two important considerations for the design and planning of cellular networks. One effective approach to handle the tradeoff between power consumption and QoS is to switch some of the Base Stations (BSs) to sleep mode when traffic load is low. In this paper, we model each BS as a processor-sharing queue with vacations, and investigate the performance of three BS sleeping schemes, namely the isolated scheme in which each BS switches mode based on its own load, the cooperative scheme in which traffic is allowed to overflow from sleeping BSs to neighboring active BSs, and a combination of both schemes. We propose a robust, scalable and computationally efficient analytical method to evaluate QoS metrics (including mean delay and blocking probability) and power consumption for each scheme and validate their accuracy by simulations. We also demonstrate the power consumption and QoS trade-off by extensive and statistically reliable experiments, and compare the performance of the three schemes under different network conditions.

Channel borrowing has been widely adopted in mobile cellular networks as a compromise between fixed channel assignment and dynamic channel assignment. Various channel borrowing schemes have been proposed in the literature with their performance, typically in terms of blocking probability, obtained by computer simulations. However, analytical approximation methods are much more computationally effective and practically useful for performance analysis of mobile cellular networks. In this paper, we model mobile cellular networks with channel borrowing as overflow loss systems. We present two methods for approximation of blocking probabilities in such systems, one based on the classical Erlang fixed point approximation (EFPA) approach, and the other derived from the recently established information exchange surrogate approximation (IESA) framework. Comprehensive results of both EFPA and IESA are provided and compared against simulation results. We demonstrate for a wide range of scenarios that IESA is more accurate in evaluating blocking probabilities than EFPA. To the best of our knowledge, our proposed approximation framework is the first workable approach in terms of accuracy and computational efficiency for cellular networks with consideration to mobility of calls and a channel borrowing mechanism.

Abstract—I/O performance is vital for most HPC applications especially those that generate a vast amount of data with the growth of scale. Many studies have shown that scientific applications tend to issue small and noncontiguous accesses in an interleaving fashion, causing different processes to access overlapping regions. In such scenario, collective I/O is a widely used optimization technique. However, the use of collective I/O deployed in existing MPI implementations is not trivial and sometimes even impossible. Collective I/O is an optimization based on a single collective I/O access. If the data reside in different places (e.g. in different arrays), the application has to maintain a buffer to first combine these data and then perform I/O operations on the buffer rather than the original data pieces. The process is very tedious for application developers. Besides, collective I/O requires the creating of a file view to describe the

To effectively model various regions with different resolutions, adaptive mesh refinement (AMR) is commonly used in cosmology simulations. There are two well-known numerical approaches towards the implementation of AMR based cosmology simulations: block-based AMR and cell-based AMR. While many studies have been conducted to improve performance and scalability of block-structured AMR applications, little work has been done for cell-based simulations. In this study, we present a parallel IO design for cell-based AMR cosmology applications, in particular, the ART(Adaptive Refinement Tree) code. First, we design a new data format that incorporates a space filling curve to map between spatial and on-disk locations. This indexing not only enables concurrent IO accesses from multiple application processes, but also allows users to extract local regions without significant additional memory, CPU or disk space overheads. Second, we develop a flexible N-M mapping mechanism to harvest the benefits of N-N and N-1 mappings where N is number of application processes and M is a user-tunable parameter for number of files. It not only overcomes the limited bandwidth issue of an N-1 mapping by allowing the creation of multiple files, but also enables users to efficiently restart the application at a variety of computing scales. Third, we develop a user-level library to transparently and automatically aggregate small IO accesses per process to accelerate IO performance. We evaluate this new parallel IO design by means of real cosmology simulations on production HPC system at TACC. Our preliminary results indicate that it can not only provide the functionality required by scientists (e.g., effective extraction of local regions and flexible process-to file mapping), but also significantly improve IO performance.

Cosmology simulations are highly communication-intensive, thus it is critical to exploit topology-aware task mapping techniques for performance optimization. To exploit the architectural properties of multiprocessor clusters (the performance gap between inter-node and intra-node communication as well as the gap between inter-socket and intra-socket communication), we design and develop a hierarchical task mapping scheme for cell-based AMR (Adaptive Mesh Refinement) cosmology simulations, in particular, the ART application. Our scheme consists of two parts: (1) an inter-node mapping to map application processes onto nodes with the objective of minimizing network traffic among nodes and (2) an intra-node mapping within each node to minimize the maximum size of messages transmitted between CPU sockets. Experiments on production supercomputers with 3D torus and fat-tree topologies show that our scheme can significantly reduce application communication cost by up to 50%. More importantly, our scheme is generic and can be extended to many other applications.

Cosmological simulations are highly complicated, and it is time-consuming to redesign and reimplement the code for improvement. Moreover, it is a risk to implement any idea directly in the code without knowing its effects on performance. In this paper, we design an emulator for cell-based AMR (adaptive mesh refinement) cosmology simulations, in particular, the Adaptive Refinement Tree (ART) application. ART is an advanced "hydro+N-body" simulation tool integrating extensive physics processes for cosmological research. The emulator is designed based on the behaviors of cell-based AMR cosmology simulations, and quantitative performance models are built toward the design of the emulator. Our experiments with realistic cosmology simulations on production supercomputers indicate that the emulator is accurate. Moreover, we evaluate and compare three different load balancing schemes for cell-based cosmology simulations via the emulator. The comparison results provide us useful insight into the performance and scalability of different load balance schemes.