SVMSched (Smart Virtual Machine Scheduler)
brief description:
SVMSched is a VM orchestrator enabling to set up on-demand- SaaS and PaaS clouds. Relying on OpenNebula, it allows to schedule requests through "Smart Virtual Machines" that allow to configure execution environments automatically.
opennebula version:
1.4.x, 2.x
Rodrigue Chakode
{{ ecosystem:logo_INRIA.png }}
last update:

SVMSched (Smart Virtual Machine Scheduler) is a tool enabling and easing the set-up of on-demand SaaS and PaaS cloud infrastructures on top of OpenNebula. Relying on an high-level abstraction of virtual machines, along with a high-capability of dynamicity and customization introduced in the provisioning of virtual machines it enables powerful and flexible resource management capabilities in scheduling the execution of requests.


- Configuration-based On-demand Cloud Services: SVMSched allows you to define a set of services or applications that will be provided on a virtual infrastructure. All services are defined in a single configuration file (an XML file), each service having a set of parameters (e.g. paths to context data, a remote data repository) that are used to dynamically create virtual machines to run service requests on-demand.

- Automatic provisioning and high-level abstraction of virtual machines: After deploying a SVMSched-based cloud, you can run services by making simple requests as for example “I want a VM with 4 CPUs and 512 MB of memory to compute a given set of data with a given application (defined in the configuration file)”. Then, SVMSched will do the rest for you (prepare the VM's image, instantiate the VM, deploy and start the VM on a node it selected seamlessly, start the execution of the service within the VM, shut down the VM when the execution is completed).

- Scheduling: SVMSched enables advanced scheduling policies such as, task prioritization, best-effort along with automatic preemption and resuming (plus migration, where required), resource sharing, etc.

- Remote Data Repository: SVMSched is designed so to enable you to define shared data repositories on the network that can be mounted automatically within the file system of virtual machines at startup. Such a repository can be actually used to store data so to ease the providing of input data as well as the retrieving of output data. Hence, you can minimize the risk of losing data and computation already done if a virtual machine fails unexpectedly.

- Learn more about SVMSched here :

ecosystem:svmsched · Last modified: 2014/01/07 17:03 by
Admin · Login