Subscribe to the gold package and get unlimited access to Shamra Academy
Register a new userA Study on the Splitting Strategy of Server Resources
52
0
0.0
(
0
)
Added by
Yiheng Shen
Publication date
2020
fields
Informatics Engineering
and research's language is
English
Authors
Yiheng Shen
Ask ChatGPT about the research
No Arabic abstract
The paper is based on Noars model of charged queues. We extend this model into multi-server systems with information about length and service rate disclosed for all the customers, and the customers can choose the optimal options. We discuss whether the splitting strategy of the server resource could bring more revenue for the service provider. We prove that any G/D/1 server supplier cannot earn more revenue by splitting his resource under the equal-toll limitations.
rate research
Read More
We introduce a system-wide safety staffing (SWSS) parameter for multiclass multi-pool networks of any tree topology, Markovian or non-Markovian, in the Halfin-Whitt regime. This parameter can be regarded as the optimal reallocation of the capacity fluctuations (positive or negative) of order $sqrt{n}$ when each server pool employs a square-root staffing rule. We provide an explicit form of the SWSS as a function of the system parameters, which is derived using a graph theoretic approach based on Gaussian elimination. For Markovian networks, we give an equivalent characterization of the SWSS parameter via the drift parameters of the limiting diffusion. We show that if the SWSS parameter is negative, the limiting diffusion and the diffusion-scaled queueing processes are transient under any Markov control, and cannot have a stationary distribution when this parameter is zero. If it is positive, we show that the diffusion-scaled queueing processes are uniformly stabilizable, that is, there exists a scheduling policy under which the stationary distributions of the controlled processes are tight over the size of the network. In addition, there exists a control under which the limiting controlled diffusion is exponentially ergodic. Thus we have identified a necessary and sufficient condition for the uniform stabilizability of such networks in the Halfin-Whitt regime. We use a constant control resulting from the leaf elimination algorithm to stabilize the limiting controlled diffusion, while a family of Markov scheduling policies which are easy to compute are used to stabilize the diffusion-scaled processes. Finally, we show that under these controls the processes are exponentially ergodic and the stationary distributions have exponential tails.
Optimization in distributed networks plays a central role in almost all distributed machine learning problems. In principle, the use of distributed task allocation has reduced the computational time, allowing better response rates and higher data reliability. However, for these computational algorithms to run effectively in complex distributed systems, the algorithms ought to compensate for communication asynchrony, network node failures and delays known as stragglers. These issues can change the effective connection topology of the network, which may vary over time, thus hindering the optimization process. In this paper, we propose a new distributed unconstrained optimization algorithm for minimizing a convex function which is adaptable to a parameter server network. In particular, the network worker nodes solve their local optimization problems, allowing the computation of their local coded gradients, which will be sent to different server nodes. Then within this parameter server platform each server node aggregates its communicated local gradients, allowing convergence to the desired optimizer. This algorithm is robust to network s worker node failures, disconnection, or delaying nodes known as stragglers. One way to overcome the straggler problem is to allow coding over the network. We further extend this coding framework to enhance the convergence of the proposed algorithm under such varying network topologies. By using coding and utilizing evaluations of gradients of uniformly bounded delay we further enhance the proposed algorithm performance. Finally, we implement the proposed scheme in MATLAB and provide comparative results demonstrating the effectiveness of the proposed framework
We consider the problem of dividing limited resources between a set of agents arriving sequentially with unknown (stochastic) utilities. Our goal is to find a fair allocation - one that is simultaneously Pareto-efficient and envy-free. When all utilities are known upfront, the above desiderata are simultaneously achievable (and efficiently computable) for a large class of utility functions. In a sequential setting, however, no policy can guarantee these desiderata simultaneously for all possible utility realizations. A natural online fair allocation objective is to minimize the deviation of each agents final allocation from their fair allocation in hindsight. This translates into simultaneous guarantees for both Pareto-efficiency and envy-freeness. However, the resulting dynamic program has state-space which is exponential in the number of agents. We propose a simple policy, HopeOnline, that instead aims to `match the ex-post fair allocation vector using the current available resources and `predicted histogram of future utilities. We demonstrate the effectiveness of our policy compared to other heurstics on a dataset inspired by mobile food-bank allocations.
The concept of intelligent reflecting surfaces (IRSs) is considered as a promising technology for increasing the efficiency of mobile wireless networks. This is achieved by employing a vast amount of low-cost individually adjustable passive reflect elements, that are able to apply changes to the reflected signal. To this end, the IRS makes the environment realtime controllable and can be adjusted to significantly increase the received signal quality at the users by passive beamsteering. However, the changes to the reflected signals have an effect on all users near the IRS, which makes it impossible to optimize the changes to positively influence every transmission, affected by the reflections. This results in some users not only experiencing better signal quality, but also an increase in received interference. To mitigate this negative side effect of the IRS, this paper utilizes the rate splitting (RS) technique, which enables the mitigation of interference within the network in such a way that it also mitigates the increased interference caused by the IRS. To investigate the effects on the overall power savings, that can be achieved by combining both techniques, we minimize the required transmit power, needed to satisfy per-user quality-of-service (QoS) constraints. Numerical results show the improved power savings, that can be gained by utilizing the IRS and the RS technique simultaneously. In fact, the concurrent use of both techniques yields power savings, which are beyond the cumulative power savings of using each technique separately.
We develop and analyze a measure-valued fluid model keeping track of parking and charging requirements of electric vehicles in a local distribution grid. We show how this model arises as an accumulation point of an appropriately scaled sequence of stochastic network models. The invariant point of the fluid model encodes the electrical characteristics of the network and the stochastic behavior of its users, and it is characterized, when it exists, by the solution of a so-called Alternating Current Optimal Power Flow (ACOPF) problem.
Log in to be able to interact and post comments
comments
Fetching comments
Sign in to be able to follow your search criteria
