Analyzing processing overhead of Type-0 hypervisor for cloud gaming

The recent advancement in hardware virtualization has allowed the possibility of Gaming as a Service cloud development. Though an ambitious and promising technology, the input latency and inability to provide decent payable frames per second have led to the failures of organization such as OnLive trying to provide cloud gaming. But with recent close to metal support of graphics API such as DirectX 12 and Vulkan and Sony is providing PS3 games via its PS Now service, cloud gaming seems to be back in the market. Still the hypervisor area is a major place where we haven't seen much improvement other than vSGA and vDGA. This research tries to analyze the various available Type-0 hypervisor in hope to find the most optimal option. The work proposes comparing three type-0 hypervisors namely KVM, Hyper-V and ESXi. The performance will be measured by various graphics benchmarking tools and the FPS in various current generation games. Various benchmarking tools are used to make a fair basis for comparison of the hypervisors.

[1]  Michael Duelli,et al.  Performance Comparison of Common Server Hardware Virtualization Solutions Regarding the Network Throughput of Virtualized Systems , 2011 .

[2]  Nadia Hocine,et al.  Network Aware Traffic Adaptation for Cloud Games , 2013, 2013 International Conference on Cloud Computing and Big Data.

[3]  Kajal T. Claypool,et al.  Latency can kill: precision and deadline in online games , 2010, MMSys '10.

[4]  Ryan Shea,et al.  Cloud gaming: architecture and performance , 2013, IEEE Network.

[5]  Timothy Wood,et al.  A component-based performance comparison of four hypervisors , 2013, 2013 IFIP/IEEE International Symposium on Integrated Network Management (IM 2013).

[6]  John Paul Walters,et al.  A Comparison of Virtualization Technologies for HPC , 2008, 22nd International Conference on Advanced Information Networking and Applications (aina 2008).

[7]  Diogo Gomes,et al.  Benchmarking the Performance of Hypervisors on Different Workloads , 2014, ICSEA 2014.

[8]  Kai Hwang,et al.  Game cloud design with virtualized CPU/GPU servers and initial performance results , 2012, ScienceCloud '12.

[9]  Abdelkader Gouaïch,et al.  Network Traffic Adaptation For Cloud Games , 2013, CloudCom 2013.

[10]  Chun-Ying Huang,et al.  Measuring the latency of cloud gaming systems , 2011, ACM Multimedia.

[11]  Jeanna Neefe Matthews,et al.  Quantifying the performance isolation properties of virtualization systems , 2007, ExpCS '07.