sakergrid w/ MiCADOscale

Dynamic Allocation of VMs for deadline-based scaling

SakerGrid is a server-app to plan and execute multiple parameter sweep simulations. To further optimize the execution time of the simulations, MiCADOscale is used to perform deadline-based scaling through dynamic allocation of Virtual Machines. This resulted in faster decision-making/runtime with optimised costs.

scaling metricstime to deadlinequeue lengthjobs runningjobs failed/completedtime elapsedaverage job durationfar-sighted decisionseconomic benefitsraising awareness on decisions, threats, impactsfaster decision makingawareness supports validation of decisionsoptimization of costsideal allocation of compute resourcesfast as a rocket order processinglaunch and flight control for optimal perfomancesakergrid w/ MiCADOSoftwaresakergridDeveloped bySaker Solutions Ltd.Planning and execution of multiple parameter sweep simulationsOpen-Source noserver applicationbottleneckexecution timescaling_levelvirtual instancesprogram_typescale

Dynamic Allocation of VMs for deadline-based scaling

SakerGrid w/ MiCADOscale is an industrial demonstrator of the COLA Project (2016-2019).

The COLA Project – Cloud Orchestration at the Level of Application (COLA) received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No 731574.