What Is Cloud Computing

Since telecom networks must be always on, it is critical to guarantee network service availability with the minimum amount of resources. In this paper, we presented XXX, our approach to provisioning availability in NFV. XXX provides a suite for service providers to concisely describe a data center topology and estimate the downtime distribution of a mapped SFC request. To reduce the running time of XXX, we further proposed an optimization framework which estimates the minimum amount of backup resources required to meet heterogeneous availability requirements and showed an heuristic solution with a theoretical bound. Through evaluation with data from production data centers, we demonstrated the effectiveness and efficiency of our solutions. For the future work, we are expanding our optimization framework to enable expedited estimations of the backup resource needed under other models of VNF. On the other hand, the downtime estimation accuracy is heavily dependent on the accuracy of the failure and repair distribution of each layer, including both hardware and software. Therefore it is important to analyze system log data, extract and model them using appropriate general

There are multiple technologies can help address the bandwidth and server lag time issues affecting users in the scenario. These issues can largely be alleviated by the addition of the load balancer and added Server virtual Machines (VMs). The load

The Important of Cloud Computing is increasing and it is receiving a growing attention in Scientific & Industrial Communities. Cloud computing is one of the top 10 most important technologies and has a greater possibility in successive years by companies and organizations. Cloud computing implement everywhere, favourable, on demand network access to a shared pool of configurable computing resources (e.g. networks, servers, storage, applications and servers) that can be immediately supply and released with minimal management effort.

Various virtualized platform that helps user to accomplish their jobs with minimum response time and minimum completion time that are provided by cloud computing environment. There are three types of services in the cloud, they are Infrastructure as a Service (Iaas), Platform as a Service (Paas), Software as a service (SaaS). The characteristics of cloud computing are resources polling, elasticity, on demand services, broad network access and middleware capabilities from provider of delivering Iaas and Paas in the form of datacenter, above services provide by deployment models such as Public cloud, Private cloud, community cloud and hybrid cloud that are used by the cloud provider. Job Scheduling has the main goal to achieve a high performances computing and best system throughput. The important issues in scheduling are Priority of job because some jobs should be serviced earlier than other those jobs cannot wait for a long time in the system. Priority of jobs must be considered by a suitable job scheduling algorithm. To address this problem some of the priority of jobs scheduling algorithm must be considered. The role of Virtual Machine’s (VMs) is an important issue because, through virtualization technology, it makes the infrastructure of cloud computing to be scalable. Therefore the important issue is to develop on optimal scheduling of virtual machine.

Schedule computation or schedule communication helps to optimize utilization and keep running time low. Several works propose to improve job scheduling by preserving data locality maintaining fair allocation among multiple resource types or discarding time-consuming tasks. Even with optimal computation scheduling, the cluster network can still become a blockage. The optimization of network transfers can be done by improving the flow bandwidth allocation or by dynamically changing paths in response to demand. These approaches need accurate and timely application demand information, obtained either from the application itself through instrumentation, which is quick and accurate but intrusive, or from the network through monitoring , which does not require application involvement, but can be expensive, slow, and detects changes in demand only after they have occurred.

Chase, J., & Doyle, R. (2001, May). Balance of power: Energy management for server clusters. In Proceedings of the 8th Workshop on Hot Topics in Operating Systems (HotOS) (Vol. 200, No. 1).

Abstract— the great emergence of mobile cloud computing has gained lots of interest in these recent years. Mobile cloud computing consists of mobile devices and cloud computing offering mobile cloud environment by using a huge number of different mobile devices. Any device uses mobility such as laptops, tablets, and smart phones is considered as a mobile device. Due to the great improvement of wireless networks like 4G and the expansion of use of mobile device, many researchers are interested to provide different methods to use mobile resources for distributed processes to compute many applications in mobile cloud environment. Although the last mobile devices have improved processing power, there are some limitations in power capacity, unpredictable network conditions, and connections availability that must be considered in order to use mobile devices as resource. These dynamic changes may disconnect a mobile device from the cloud computing and then prevent applications from completion. In other words, these problems may cause a system to fault more frequently due to these changes. Therefore, in order to increase the stability of these devices, there should be a monitoring technique to tolerate system faults. In this paper, we focus on Markov Chain, which precisely collects and analyzes the information of states and proactively predicts the status of future state in order to solve the volatility and instability of mobile devices. Also, we study two fault tolerant methods,

Cloud computing was founded on already-existing computing technologies, such as distributed computing (Clusters, Grids), Internet technologies (Web Services, Service Oriented Architectures, Web 2.0), virtualization and systems management (autonomic computing, data centre automation) [11]. Specifically, with the advent of Cloud computing, the benefits of these technologies have been combined and further improved. Cloud basically presents a suitable environment for convenient, on-demand network access to a pool of configurable computing resources that can be mutually accessed and provisioned with insignificant administration effort or service provider interaction [1]. The optimal provisioning of these compute resources to meet certain predefined performance criteria is a complex problem which has attracted much research attention [12][13]. This problem comprise the steps involved in the allocation of suitable computing resources to tasks with the aim of optimizing certain objectives. Some objective functions include but not limited to; maximizing task throughput, minimizing the time of task completion and maximizing the utilization of resources.

Without including any network-specific functionality, it would be possible to develop a scalable workload management system. While the connectivity between compute nodes is required and also for the external access, it is important to have network-specific functionality, but it would be possible to benefit from the existing networking infrastructure to allocate IP addresses and relay data between nodes. However, the main problem with such an approach is the network management system would not be able to separate traffic between users securely and efficiently when there will be a multitenant environment[13]. Neutron is a part of OpenStack project focused on delivering networking as a service and manages the

While High Performance Cloud Computing model provides more flexibility, it faces several challenges: Sometimes companies have data and applications which consider as their property and they may not want to send it to shared resources. So some cloud providers offer special multi-tenant environments that is called virtual private clouds. Although it is hosted by the provider, customer has a full control on it. Some services offerings even go for shared-nothing resources to be exclusively used by a critical application [5].

The final steps include the analysis of the developed system and report to the authorities.

To cut down energy utility it is mandatory to join the other workloads. This piece of article is efficiently able to explain how with great efficiency heterogeneous workloads can be managed and s ale for standardization is Single threshold technique. Concentric attention in this technique is that results explains how this method is well-efficient for the data centres to join the multigeneous workloads and energy is consumed by varied data centres due to heterogeneous applications and finds out SLA violations with QOS.

Cloud computing is one of the new models of computing that is used to provide for the computing infrastructure. A good computing cloud data management system must satisfy the following goals like availability, scalability, elasticity, performance [1]. The cloud computing looks to satisfy the needs of various levels of clients and workloads. Hence a guaranteed contract between the service provider and the clients has been made known as the Service Level Agreement (SLA). Both the providers and consumers consider successful SLA management.

New DC/OS edge load-balancer adds unique, customizable and high-performance inbound traffic management to container-based environments.

This paper does a survey on the current dynamic load balancing strategies to meet the need of cloud computing environments. This survey starts with an overview of each of these strategies and approaches, then describes the properties and finally compares them based on multiple criteria like : Resource Utilization, Traffic Management, Throughput, Response Before getting into the challenges and different cloud computing techniques, few areas like its architecture and how it differs from traditional data centers, need to be discussed. Now conventionally a cloud system is divided into two components - front-end and backend.[29]. Both of these two components are generally connected with each other via internet. As the name suggests, Front end is what a user sees and on the other hand,