Computing clouds are widely used by many organizations in science, business, and industry. They provide flexible access to various physical computing resources. In addition, computing clouds allow better resource utilization. Today many scientific organizations have their own private cloud used for both hosting services and performing computations. In many cases, private clouds are not 100% loaded and may be used as work nodes for distributed computations. If one connects clouds together, it will be possible to use them together for computational tasks. Therefore, the idea to integrate several private clouds appeared. A cloud bursting approach can be used for the integration of resources. However, in order to provide access to the united cloud for all participants, an extensive configuration will be required in all clouds. We studied the possibility to combine clouds by integrating them using a distributed workload management system – DIRAC Interware. Two approaches to spawning virtual machines were evaluated: the use of the OCCI interface and the native OpenNebula XML-RPC interface. They were tested, and both approaches allowed performing computing jobs in various clouds based on the OpenNebula software.
|Number of pages||5|
|Journal||CEUR Workshop Proceedings|
|State||Published - 1 Jan 2019|
|Event||27th Symposium on Nuclear Electronics and Computing, NEC 2019 - Budva, Montenegro|
Duration: 30 Sep 2019 → 4 Oct 2019
Scopus subject areas
- Computer Science(all)