作者: Mansoor Alicherry , Ashok Anand , Shoban Preeth Chandrabose , Theophilius Benson
关键词: Engineering 、 Operating system 、 Provisioning 、 Cloud computing 、 Legacy system 、 Granularity 、 Web server 、 Sizing 、 Downtime 、 Virtual machine
摘要: The current IaaS model has several shortcomings. First, providers only offers VM (virtual machine) with predefined sizes, thus enterprise tenants must judiciously determine the size that best fit their application. This is challenging as overprovisioning VMs can lead to waste of resources while underprovisioned poor performance. Second, when an application requires more than a provide, are currently limited either scaling-out or scaling-up applications. However, in both situations granularity at level which leads sizing issues discussed earlier. Third, ineffective it incurs significant amount downtime/poor performance new being provisioned and not all applications support scaling-out. For example while, Web servers be easily scaled-out other legacy [1], limiting its applicability.