Resource Placement in Cloud Computing and Network Applications
Resource Placement in Cloud Computing and Network Applications
复制标题
DOI：
--
复制
发表时间：
2017
影响因子：
3.6
通讯作者：
Yuval Rochman
中科院分区：
文献类型：
--
作者：
Yuval Rochman
在这篇论文中，我们解决的问题，资源分配在地理分布的系统，特别是云计算。我们认为，一个大规模的系统和区域分布的各种资源的多种类型的需求。系统运营商的目标是跨区域配置资源以实现利润最大化，因此需要解决如何根据需求配置资源的问题。该系统面临的是一个任意的，多维的，随机的需求，因此，运营商应优化利润，同时考虑到充分的需求分布。云计算和在线服务，利用区域数据中心，并面临在哪里放置各种服务器的问题，属于这种模式。我们的问题是在一个非常广泛的成本模型下发展起来的，因此可以容纳许多系统。这些问题根据它们所代表的系统有各种各样的公式。我们提供高效的算法运行在一个快速的多项式时间1。我们的算法和解决方案可用于：1.精确的系统优化; 2.推导出基于启发式分析的下限; 3.敏感性分析。我们的解决方案是基于分析技术，利用随机分析以及图论方法，可以应用到其他优化/组合问题。1操作员放置或重新定位的资源数量的多项式。III
In this thesis we address the problem of resource allocation in geographically distributed systems, in particular cloud computing. We consider a large-scale system and regionally distributed demands for various resources of multiple types. The system operator aims at placing the resources across regions to maximize the profit, thus needing to address the problem of how to place the resources in response to the demand. The system is faced with an arbitrary, multi-dimensional, stochastic demand, thus the operator should optimize the profit while taking into account the full demand distributions. Cloud computing and online services, utilizing regional datacenters and facing the problem of where to place various servers, fall under this paradigm. Our problems are developed under a very wide cost model, thus allowing the accommodation of many systems. These problems have various formulations based on the systems they represent. We provide efficient algorithms running in a fast polynomial time1. Our algorithms and solutions can be used for: 1. Exact system optimization; 2. Deriving lower bounds for heuristic based analysis, and; 3. Sensitivity analysis. Our solutions are based on analytic techniques utilizing stochastic analysis as well as graph theory methodologies that can be applied to other optimization/combinatorial problems. 1Polynomial in the number of resources the operator places or reposition. iii