3 GPP LTE has approved uplink intra-cell power control and defined overload indicator （OI） for uplink inter-cell power contrQ1 to mitigate the inter-cell interference （ICI）, respectively. In this pa- per, we propo...3 GPP LTE has approved uplink intra-cell power control and defined overload indicator （OI） for uplink inter-cell power contrQ1 to mitigate the inter-cell interference （ICI）, respectively. In this pa- per, we propose a hierarchical power control （ HPC ） scheme where intra-eell and inter-cell power controls interact with each other. The inter-cell power control eommand is generated by radio re- source management （RRM） entity according to the ICI-load model together with the current ICI and served load information. This ICI-load model is proposed as a guideline for coordination among cells to enable the system to approach its system specific interference over thermal noise （IoT） work area. Simulation results show that for HPC scheme, the system＇ s IoT is well controlled to fit its pre-de- fined work area and the power efficiency is improved significantly. Our proposed scheme is also ro- bust to different settings of its inter-cell power control period.展开更多
Modern computer systems are increasingly bounded by the available or permissible power at multiple layers from individual components to data centers.To cope with this reality,it is necessary to understand how power bo...Modern computer systems are increasingly bounded by the available or permissible power at multiple layers from individual components to data centers.To cope with this reality,it is necessary to understand how power bounds im-pact performance,especially for systems built from high-end nodes,each consisting of multiple power hungry components.Because placing an inappropriate power bound on a node or a component can lead to severe performance loss,coordinat-ing power allocation among nodes and components is mandatory to achieve desired performance given a total power bud-get.In this article,we describe the paradigm of power bounded high-performance computing,which considers coordinated power bound assignment to be a key factor in computer system performance analysis and optimization.We apply this paradigm to the problem of power coordination across multiple layers for both CPU and GPU computing.Using several case studies,we demonstrate how the principles of balanced power coordination can be applied and adapted to the inter-play of workloads,hardware technology,and the available total power for performance improvement.展开更多
基金Supported by the National High Technology Research and Development Programme of China(No.2009AA011501)International S&T Cooperation Program of Shanghai Municipality(No.09530702500&10220712100)Major Project of Shanghai Municipality(No.09511501100)
文摘3 GPP LTE has approved uplink intra-cell power control and defined overload indicator （OI） for uplink inter-cell power contrQ1 to mitigate the inter-cell interference （ICI）, respectively. In this pa- per, we propose a hierarchical power control （ HPC ） scheme where intra-eell and inter-cell power controls interact with each other. The inter-cell power control eommand is generated by radio re- source management （RRM） entity according to the ICI-load model together with the current ICI and served load information. This ICI-load model is proposed as a guideline for coordination among cells to enable the system to approach its system specific interference over thermal noise （IoT） work area. Simulation results show that for HPC scheme, the system＇ s IoT is well controlled to fit its pre-de- fined work area and the power efficiency is improved significantly. Our proposed scheme is also ro- bust to different settings of its inter-cell power control period.
基金supported in part by the U.S.National Science Foundation under Grant Nos.CCF-1551511 and CNS-1551262.
文摘Modern computer systems are increasingly bounded by the available or permissible power at multiple layers from individual components to data centers.To cope with this reality,it is necessary to understand how power bounds im-pact performance,especially for systems built from high-end nodes,each consisting of multiple power hungry components.Because placing an inappropriate power bound on a node or a component can lead to severe performance loss,coordinat-ing power allocation among nodes and components is mandatory to achieve desired performance given a total power bud-get.In this article,we describe the paradigm of power bounded high-performance computing,which considers coordinated power bound assignment to be a key factor in computer system performance analysis and optimization.We apply this paradigm to the problem of power coordination across multiple layers for both CPU and GPU computing.Using several case studies,we demonstrate how the principles of balanced power coordination can be applied and adapted to the inter-play of workloads,hardware technology,and the available total power for performance improvement.
