To reduce energy consumption in cloud data centres,in this paper,we propose two algorithms called the Energy-aware Scheduling algorithm using Workload-aware Consolidation Technique(ESWCT) and the Energyaware Live Migr...To reduce energy consumption in cloud data centres,in this paper,we propose two algorithms called the Energy-aware Scheduling algorithm using Workload-aware Consolidation Technique(ESWCT) and the Energyaware Live Migration algorithm using Workload-aware Consolidation Technique(ELMWCT).As opposed to traditional energy-aware scheduling algorithms,which often focus on only one-dimensional resource,the two algorithms are based on the fact that multiple resources(such as CPU,memory and network bandwidth)are shared by users concurrently in cloud data centres and heterogeneous workloads have different resource consumption characteristics.Both algorithms investigate the problem of consolidating heterogeneous workloads.They try to execute all Virtual Machines(VMs) with the minimum amount of Physical Machines(PMs),and then power off unused physical servers to reduce power consumption.Simulation results show that both algorithms efficiently utilise the resources in cloud data centres,and the multidimensional resources have good balanced utilizations,which demonstrate their promising energy saving capability.展开更多
Graph is a significant data structure that describes the relationship between entries.Many application domains in the real world are heavily dependent on graph data.However,graph applications are vastly different from...Graph is a significant data structure that describes the relationship between entries.Many application domains in the real world are heavily dependent on graph data.However,graph applications are vastly different from traditional applications.It is inefficient to use general-purpose platforms for graph applications,thus contributing to the research of specific graph processing platforms.In this survey,we systematically categorize the graph workloads and applications,and provide a detailed review of existing graph processing platforms by dividing them into generalpurpose and specialized systems.We thoroughly analyze the implementation technologies including programming models,partitioning strategies,communication models,execution models,and fault tolerance strategies.Finally,we analyze recent advances and present four open problems for future research.展开更多
Benchmarks play a crucial role in database performance evaluation,and have been effectively promoting the development of database management systems.With critical transaction processing requirements of new application...Benchmarks play a crucial role in database performance evaluation,and have been effectively promoting the development of database management systems.With critical transaction processing requirements of new applications,we see an explosion of innovative database technologies for dealing with highly intensive transaction workloads(OLTP)with the obvious characteristics of sharp dynamics,terrificskewness,high contention,or high concurrency(abbr.DSC^2),which can not be well described or evaluated by current standard benchmarks.In this paper,based on the representative SecKill applications,we define a pacakge of workloads simulating intensive transactional processing requirements.And we create a general and flexible benchmark framework PeakBench for evaluating intensive OLTP workloads on databases.We are the first work to have full control on simulating DSC^2,especially for the fine granularity control for contention generation.With a comprehensive set of experiments conducted on popular open sourced DBMSs compared with the other representative OLTP benchmarks,we completely demonstrate the usefulness of PeakBench.展开更多
基金supported by the Opening Project of State key Laboratory of Networking and Switching Technology under Grant No.SKLNST-2010-1-03the National Natural Science Foundation of China under Grants No.U1333113,No.61303204+1 种基金the Sichuan Province seedling project under Grant No.2012ZZ036the Scientific Research Fund of Sichuan Normal University under Grant No.13KYL06
文摘To reduce energy consumption in cloud data centres,in this paper,we propose two algorithms called the Energy-aware Scheduling algorithm using Workload-aware Consolidation Technique(ESWCT) and the Energyaware Live Migration algorithm using Workload-aware Consolidation Technique(ELMWCT).As opposed to traditional energy-aware scheduling algorithms,which often focus on only one-dimensional resource,the two algorithms are based on the fact that multiple resources(such as CPU,memory and network bandwidth)are shared by users concurrently in cloud data centres and heterogeneous workloads have different resource consumption characteristics.Both algorithms investigate the problem of consolidating heterogeneous workloads.They try to execute all Virtual Machines(VMs) with the minimum amount of Physical Machines(PMs),and then power off unused physical servers to reduce power consumption.Simulation results show that both algorithms efficiently utilise the resources in cloud data centres,and the multidimensional resources have good balanced utilizations,which demonstrate their promising energy saving capability.
基金Project supported by the National Key Program of China(No.2018YFB2101100)the National Natural Science Foundation of China(Nos.61932001 and 61872376)the Major State Research Development Program of China(No.2016YFB0201305).
文摘Graph is a significant data structure that describes the relationship between entries.Many application domains in the real world are heavily dependent on graph data.However,graph applications are vastly different from traditional applications.It is inefficient to use general-purpose platforms for graph applications,thus contributing to the research of specific graph processing platforms.In this survey,we systematically categorize the graph workloads and applications,and provide a detailed review of existing graph processing platforms by dividing them into generalpurpose and specialized systems.We thoroughly analyze the implementation technologies including programming models,partitioning strategies,communication models,execution models,and fault tolerance strategies.Finally,we analyze recent advances and present four open problems for future research.
基金We are partially supported by the Key Program of National Natural Science Foundation of China(2018YFB1003402)the National Natural Science Foundation of China(Grant No.61432006).
文摘Benchmarks play a crucial role in database performance evaluation,and have been effectively promoting the development of database management systems.With critical transaction processing requirements of new applications,we see an explosion of innovative database technologies for dealing with highly intensive transaction workloads(OLTP)with the obvious characteristics of sharp dynamics,terrificskewness,high contention,or high concurrency(abbr.DSC^2),which can not be well described or evaluated by current standard benchmarks.In this paper,based on the representative SecKill applications,we define a pacakge of workloads simulating intensive transactional processing requirements.And we create a general and flexible benchmark framework PeakBench for evaluating intensive OLTP workloads on databases.We are the first work to have full control on simulating DSC^2,especially for the fine granularity control for contention generation.With a comprehensive set of experiments conducted on popular open sourced DBMSs compared with the other representative OLTP benchmarks,we completely demonstrate the usefulness of PeakBench.
