Benefits include: 1.

Dedicated server infrastructure consistent with the infrastructure supported by the worldwide Moodle community (unavailable in other hosting environments such as Amazon). 2. Service Level Agreement guaranteeing 99.99% uptime. 3. 24/7 hosting support. 4. Daily and monthly backups. 5. Ease of scalability on-demand. 6. Fully redundant infrastructure for servers and network storage. 7. Security separation. Each client has its own unique server instance. 8. More rapid development, testing and deployment of customizations. 9. Greater economy-of-scale with respect to your Moodle customization requests. 10. Cloud technology will allow provisioning of exactly the resources needed.

Cloud computing is an exciting paradigm in information systems integration in both large and small enterprises. Cloud technology will allow provisioning of exactly the resources needed. There are several types of clouds. Amazon EC2 or GoGrid are Infrastructure as a Service clouds (IaaS) offering entire instances (i.e. virtual machines) to customers. Other types are Software as a Service (SaaS) for applications (e.g., Salesforce and Google Apps ) and Platform as a Service (PaaS) clouds for development and service platforms (e.g., Google App Engine and Microsoft Azure).

Infrastructure Provider (IP). Cloud vendors that offer infrastructure (e.g., hardware) such as CPU, memory, storage, networking, etc. as a service in a pay-per use basis. One key concept of these offerings is elasticity: customers are capable of increasing/decreasing the amount of resources required depending on their needs (e.g., during peak loads more resources can be requested, while when the usage is low some resources might be released). Service Provider (SP): Companies and organizations that are interested in \renting" capacity (e.g., processing power) from cloud infrastructure providers in order to deploy their services in a highly scalable infrastructure, due to the benefits achieved compared to using in-house infrastructure.

One key element of the interaction between SPs and IPs is the service definition mechanism. In IaaS clouds this is commonly specialized by packaging the software stack (operating systems, middleware and service components) in one or more virtual machines, however each cloud IP has its own proprietary mechanism for services definition. This complicates inter operability between clouds, which puts the SP in a vendor lock-in position. For example, if one SP has prepared his/her service to be deployed in Amazon EC2 (using the proprietary Amazon Machine Image format) then changing this service to another cloud IP (e.g., GoGrid) is not straight forward as the image creation for each provider requires several different configuration steps. Therefore there is a need for standardizing this virtual application distribution in order to avoid vendor lock-in and facilitate interoperability among IaaS clouds. The Open Virtualization Format (OVF) objective is to specify a portable packaging mechanism to foster the adoption of Virtual Appliances (VA) (i.e. pre-configured software stacks comprising one or more virtual machines to provide self-contained services) as new software release and management model (e.g. through the development of virtual appliance lifecycle management tools) in a vendor and platform neutral way (i.e., not oriented to any virtual machine technology in particular).

Why Us: Creating a cloud is no easy feat. We are Industry experienced IT researchers working in cloud computing. Of course this could be deployed by your IT department unless you have lot of R&D time and spare hardware and whiz kids hanging out. Requirement: Do you have fast enough disks? Enough RAM? Fast enough network pipes? Speedy network, you cant serve up a cloud without some serious bandwidth. How do you secure a cloud? Its taken you how long to get your internal LAN as secure as possible? Todo: Open Nebula Virtualization stack in Linux usually comprise of hypervisors like Linux Containers (LXC), Kernel Based Virtual Machine (KVM), Xen etc. For managing these hypervisors, the opensource community came up with a common library called libvirt which has specific drivers that hook onto these core hypervisors and manage them. Cloud computing is one of the major areas where libvirt library is extensively used as part of the Cloud management stack. Software Simulation is a proven technique by which usage and performance of products are verified without having to spend time, effort or money on getting a real physical infrastructure setup done. Simulation helps in familiarising and verifying a product in a controlled environment thereby the users can acquire amble exeperience to deploy (and develop) more stable product. In Cloud computing, simulation based approaches offer significant benefits. It allows cloud customers to test

their services in repeatable environment and to rectify performance bottlenecks before deploying their services on real Clouds. For the Cloud vendors, simulation environments allow evaluation various resources leasing scenarios under varying load and pricing distributions. Such studies could aid the providers in optimizing the resource access cost with focus on improving profits. Considering the relevance of libvirt in Linux virtualization stack and benefits of having simulation components in Cloud, we are proposing a simulation module for libvirt. Libvirt already has a test driver that is used for testing libvirt without any hypervisor backend. We are working on to enhance the test driver approach by adding simulation APIs to libvirt.