[[en:centro:servizos:cloud|>>Back to main page]] ====== Responsible use of the Cloud ====== The Cloud makes available to researchers a **limited** set of resources: * Primary storage: 6,5 TB. * Secondary storage((Templates and ISO images)): 2 TB. * Processing power: 287 GHz. * RAM memory: 300 GB. * IP addresses: 100. It is **important to follow the next recommendations** to make sure that the resources you don't need are available to other researchers. ===== When creating a new VM ===== - Choose a //Compute offering// suitable to your needs. For most web applications //offering// **Cloud 2** will be enough. It can be modified later, so choose a conservative value and change it tomorrow if more is needed. - Add only a //Disk offering// if additional storage space is needed. This can be done at any later moment so if in doubt don't add it now. Take into account that this storage appears like an additional disk and configuration in the operating system will be required. - In //Network// choose an existing network, if possible, instead of creating a new one. You'll have to create a network when creating your first VM. Each network created consumes one ip address from the research network pool. ===== When destroying a VM ===== - VM are not really erased from servers when you destroy them, there is a "grace period" of a few hours during which the VM won't accept any other action. - Networks are not eliminated automatically when there are no VM connected to them. This means that if you destroy the last VM of a network you should go to the //Network// tab and manually erase it. - Additional storage disks of a VM are also not erased when the VM is destroyed. They must be manually erased in the //Storage// tab. ===== For existing VM ===== - If you have more than one network and don't know why it is advisable to use only one. Put all your VM connected to the same network and make it the default. Then erase the unused networks. - If a VM has more than one disk assigned and you're not using them, erase them in the //Storage// tab. - Check the //Compute offering// assigned to each VM. If too high (5 or more) make sure it is really needed. Try reducing the offering to something lower (VM must be powered off).