After the global pandemic,DaaS(desktop as a service)has become the first choice of many companies’remote working solution.As the desktops are usually deployed in the public cloud when using DaaS,customers are more co...After the global pandemic,DaaS(desktop as a service)has become the first choice of many companies’remote working solution.As the desktops are usually deployed in the public cloud when using DaaS,customers are more cost-sensitive which boosts the requirement of proactive power management.Prior researches in this area focus on virtual desktop infrastructure(VDI)session logon behavior modeling,but for the remote desktop service host(RDSH)-shared desktop pools,logoff optimization is also important.Existing systems place sessions by round-robin or in a pre-defined order without considering their logoff time.However,these approaches usually suffer from the situation that few left sessions prevent RDSH servers from being powered-off which introduces cost waste.In this paper,we propose session placement via adaptive user logoff prediction(SODA),an innovative compound model towards proactive RDSH session placement.Specifically,an ensemble machine learning model that can predict session logoff time is combined with a statistical session placement bucket model to place RDSH sessions with similar logoff time in a more centralized manner on RDSH hosts.Consequently,the infrastructure cost-saving can be improved by reducing the resource waste introduced by those RDSH hosts with very few hanging sessions left for a long time.Experiments on real RDSH pool data demonstrate the effectiveness of the proposed proactive session placement approach against existing static placement techniques.展开更多
文摘After the global pandemic,DaaS(desktop as a service)has become the first choice of many companies’remote working solution.As the desktops are usually deployed in the public cloud when using DaaS,customers are more cost-sensitive which boosts the requirement of proactive power management.Prior researches in this area focus on virtual desktop infrastructure(VDI)session logon behavior modeling,but for the remote desktop service host(RDSH)-shared desktop pools,logoff optimization is also important.Existing systems place sessions by round-robin or in a pre-defined order without considering their logoff time.However,these approaches usually suffer from the situation that few left sessions prevent RDSH servers from being powered-off which introduces cost waste.In this paper,we propose session placement via adaptive user logoff prediction(SODA),an innovative compound model towards proactive RDSH session placement.Specifically,an ensemble machine learning model that can predict session logoff time is combined with a statistical session placement bucket model to place RDSH sessions with similar logoff time in a more centralized manner on RDSH hosts.Consequently,the infrastructure cost-saving can be improved by reducing the resource waste introduced by those RDSH hosts with very few hanging sessions left for a long time.Experiments on real RDSH pool data demonstrate the effectiveness of the proposed proactive session placement approach against existing static placement techniques.
