Energy-Efficient SLA Guarantees for Virtualized GPU in Cloud Gaming

Abstract

Both power consumption and SLA guarantees are important concerns for cloud gaming. Recently, various approaches have been developed to effectively reduce GPU power consumption by making GPU run at low frequencies. However, Virtual Machines (VMs) running on the same physical GPU with virtualization technology are correlated, because the change of GPU frequencies will affect the SLA performance of all the VMs. In fact, both reducing power consumption and guaranteeing SLA should work together under the considerations of their correlations. This paper proposes a novel two-layer control architecture called EvGPU (Energy-efficient SLA guarantees for virtualized GPU) based on well-established feedback control techniques. The first control loop adopts a Proportional-Integral (PI) controller to ensure SLA guarantees, which in particular are measured at a predefined level of the Frames Per Second (FPS) for each online game. The secondary power control loop then adjusts GPU frequency through Dynamic Voltage/Frequency Scaling (DVFS) to reduce power consumption based on the current FPS achieved by the first loop. Empirical results demonstrate that the proposed solution can effectively reduce GPU power consumption, while achieving the required SLA performance in virtualized GPU for cloud gaming.

Formally modeling and analyzing cost‐aware job scheduling for cloud data center

“…Arianyan et al propose a holistic cloud resource management procedure based on a multicriteria decision‐making method for both determination of underloaded hosts and placement of the migrating virtual machine. A two‐layer control architecture is used to virtualize GPU (EvGPU) on the basis of well‐established feedback control techniques, whereas the work of Celaya and Sakellariou contributes a policy that adaptively decides when to switch servers on and off under a workload of parallel jobs. Its objective is to minimize both the energy consumption and the number of SLA violations.…”

Energy Conservation Using Dynamic Voltage Frequency Scaling for Computational Cloud

“…Two-layer control architecture called EvGPU (energy-efficient SLA guarantees virtualized GPU) is devised by Guan et al [ 6 ] for cloud gaming. At the first layer of the control architecture the author has implemented Proportional-Integral (PI) controller which guarantees SLA and measures the number of Frames Per Second (FPS) for each online game based on the set threshold level.…”

Towards pervasive and mobile gaming with distributed cloud infrastructure

“…The research showed the possibility of scaling a single CPU/GPU resource for multiple applications when the frames-per-second (FPS) targets are pre-defined as "just good enough". Guan et al [38] proposed a similar approach from a different perspective. Their architecture aims to minimize the energy consumption of a cloud gaming service by providing a performance feedback from the game, The concept of virtualization has been introduced already in the 1960s when IBM used it to partition a large mainframe into several logical instances [64].…”