Using network topology, transmission line parameters, transformer parameters, generator location and limits, and load location and compensation, the load-flow calculation can provide voltage magnitudes and angles for all nodes and loading of network components, such as cables and transformers. With this information, compliance to operating limitations such as those stipulated by voltage ranges and maximum loads, can be examined. This is, for example, important for determining the transmission capacity of underground cables, where the influence of cable bundling on the load capability of each cable has to be taken also into account. Due to the ability to determine losses and reactive-power allocation, load-flow calculation also supports the planning engineer in the investigation of the most economical operation mode of the network.
What is claimed is: A computer-implemented system, comprising: The computer-implemented system of claim 1, wherein the number of servers needed is based upon two ratios, one ratio between total login rate of the cluster and a maximum login rate per server and the other ratio between total number of connections of the cluster and a maximum number of connections per server.
The computer-implemented system of claim 2, wherein each ratio is multiplied by an appropriate multiplicative factor to include the additional margin. The computer-implemented system of claim 3, wherein the number of servers needed is a next highest integer value greater than or equal to a larger product between the two ratios and the appropriate multiplicative factors.
The computer-implemented system of claim 1, wherein the provisioning component is configured to schedule one or more servers to be turned on from a set of inactive servers. The computer-implemented of claim 1, wherein the provisioning component is configured to schedule one or more servers to be shut down from a set of active servers.
The computer-implemented system of claim 6, further comprising a starvation component configured to drain servers scheduled to be turned off of active connections prior to shut down. The computer-implemented system of claim 7, wherein the starvation component is configured to drain the servers until a predetermined period of time expires.
The computer-implemented system of claim 8, wherein the period of time is two hours. The computer-implemented system of claim 1, the forecast component further comprising a load skewing component configured to attempt to route new login requests to the active servers while the active servers are able to handle the new login requests, and to utilize tail servers in reserve to handle a surge in new login requests until new servers can be turned on.
The computer-implemented system of claim 10, the load skewing component further configured to, based on an upper bound on a number of connections per server and a target number of connections per server, distribute new login requests to active servers with loads less than the target number and closest to the target number.
The computer-implemented system of claim 1, the evaluation component configured to compare predicted values provided by the model component and observed values produced by the monitor component, and ascertain standard deviations of relative errors between the predicted values and the observed values to generate the forecast factors portion of the multiplicative factors.
The computer-implemented system of claim 1, the dynamic load analysis component configured to ascertain a load dispatching mechanism selected by the load dispatching component to distribute new login requests collected by the cluster. The computer-implemented system of claim 1, the dynamic load analysis component configured to calculate the dynamic factors portion of the multiplicative factors based at least in part on one or more parameters employed by a selected load dispatching mechanism.
A computer-implemented method, comprising: The computer-implemented method of claim 15, further comprising: A computer-readable storage medium storing instructions, the instructions if executed by a computing device causing the computing device to perform operations comprising: Cell phones, smart phones, personal digital assistants, computers, laptops, and other smart devices are pervasive in today's world.
Moreover, the Internet has become a global establishment that is omnipresent in the lives of many people. In addition, the Internet can be employed as a means to retrieve information retained all over the globe.
For example, such services can include, but not limited to, search engines, web mail, online chat e. Many Internet services have become integrated into everyday lives of people. To accommodate increased integration, such services are expected to scale well, to provide high performance e.
To achieve these goals, Internet services are typically deployed in clusters that include a large number of servers hosted in dedicated data centers.
A data center can house a variety of heterogeneous components such as, but not limited to, computers, storage, networking equipment. In addition, the data center includes infrastructure that distributes power to the components and provides cooling to the components.
Viewed from outside, a data center can be seen as a black box that responds to a stream of user requests via the Internet while consuming power from the electrical grid and producing waste heat. With drastic increases in demand for Internet services, data centers consume more and more resources.
The amount of resources consumed is directly related to number of hosted servers in data center as well workload of the servers. As data centers scale up to meet demand for hosted services, electricity usage skyrockets. In the United States, it is estimated that billions of kilowatt-hours are consumed by data centers; an amount sufficient to power millions of homes.
SUMMARY The following discloses a simplified summary of the specification in order to provide a basic understanding of some aspects of the specification.
This summary is not an extensive overview of the specification. It is intended to neither identify key or critical elements of the specification nor delineate the scope of the specification.
Its sole purpose is to disclose some concepts of the specification in a simplified form as a prelude to the more detailed description that is disclosed later. Dynamic provisioning relates to dynamically turning on e.
A idle server consumes a non-trivial amount of energy e. In accordance with an aspect of the subject innovation, a provisioning component is provided that determines a required number of servers in a server cluster. The provisioning component can include a forecast component that predicts login rate e.
For example, as a load distribution changes, certain servers can be under utilized and no longer required to maintain quality of service.
Further, the dynamics of a particular load dispatching scheme are considered to ensure that a proper number of servers are active to enable efficient operation of the dispatching scheme.
According to another aspect of the subject innovation, the forecast component and the dynamic load analysis component provide multiplicative factors. The multiplicative factors can be combined and employed to introduce an acceptable margin in the determined number of required servers to account for inaccuracies in forecasting and dynamic behavior of load dispatching.
The following description and the annexed drawings set forth certain illustrative aspects of the specification.View Short Term Load Forecasting Research Papers on kaja-net.com for free. Essay from insider interview judge money scholarship strategy winner winning ieee research papers on computer networks essay on co education in india voorbeeld thesis sociale wetenschappen ma thesis in english linguistics.
Short On Deductive Reasoning Deductive Reasoning Much of the confusion that surrounds the wisdom of owning your home versus renting comes from the popular notion that a home should be viewed purely as an investment, the same way you would view a stock, bond, IRA, etc.
Short papers submitted to this workshop should be limited to 4 pages while long papers should be limited to 8 pages. All papers should be formatted using the SIAM SODA macro. Authors are required to submit their papers electronically in PDF format to the submission site by pm MDT, January 19 January 31, Christiaanse, W.R.
() Short Term Load Forecasting Using Genera Exponential Smoothing. IEEE Transactions on Power Apparatus and Systems, PAS, Short-Term Load Forecasting This paper discusses the state of the art in short-term load fore- casting (STLF), that is, the prediction of the system load over an.