Atsuhiro Sawada

Energy-Aware Server Selection Algorithms in a Scalable Cluster

It is critical to reduce the electric energy consumed in information systems, especially server clusters. In this paper, we extend the multi-level power consumption (MLPC) model and the multi-level computation (MLC) model to a server with multiple CPUs. In this paper, we newly propose a totally energy-aware (TEA) algorithm to select a server for a process in a cluster. Here, servers in a cluster are first classified into subclusters. Each subcluster is characterized in terms of the electric power and computation rate. One server is randomly selected in each subcluster. Then, one server is selected so that the expected electric energy is minimum in the selected servers. We evaluate the TEA algorithm and show not only the total electric energy consumption of the servers but also the average execution time of processes are reduced in the TEA algorithm compared with other algorithms.

Energy-Aware Clusters of Servers for Storage and Computation Applications

It is now critical to reduce electric energy consumed in a cluster of servers, especially scalable systems like cloud computing systems. In clusters, most application processes like web applications use not only CPU resources but also files and databases. In this paper, we consider storage processes which read and write data in files in addition to computation processes. We propose a PCS model (power consumption model for a storage server) which shows how much electric power a server consumes to perform storage and computation processes. We also propose a CS model (a computation model for storage server) which shows how long it is expected to take to perform storage processes and computation processes. By using the PCS and CS models, we propose a local energy-aware (LEA) algorithm to select a server for a request process in a cluster so that the total electric energy consumption of the servers can be reduced. We evaluate the LEA algorithm in terms of total electric energy consumption of the servers. We show the electric energy consumed by servers to perform computation and storage processes can be reduced in the LEA algorithm.

Energy-aware Server Selection Algorithms for Storage and Computation Processes

Application processes like Web applications use not only CPU but also storages like HDD. In our previous studies, the algorithms to select a server in a cluster are proposed to energy-efficiently perform processes which use either CPU or storages. In this paper, we consider a more general type of process which does both the computation and accesses to storages. In this paper, we newly propose LEAG and GEAG algorithms to select servers to perform general processes in a cluster so that the total electric energy consumption of the servers can be reduced. We evaluate the LEAG and GEAG algorithms in terms of total electric energy consumption of the servers and average execution time of the processes. We show the electric energy consumed by servers can be reduced in the LEAG and GEAG algorithms.

Simple Energy-efficient Server Selection Algorithm in a Scalable Cluster

It is critical to reduce the electric energy consumed in server clusters. In our previous studies, the MLPCM and MLCM models are proposed with LEA and MEA server selection algorithms. Here, a server is selected to perform a process by estimating the termination time of every current process. However, it takes time to collect the state of each process and estimate the termination time of each process. In this paper, we propose a simple energy-aware (PEA) algorithm to select a server where only state information on number of processes on each server is used. In the evaluation, we show the computation complexity of the PEA algorithm is O(1), smaller than the other algorithms while the total electric energy consumption of the servers of the PEA algorithm is almost the same as the MEA algorithm and is smaller than the others.

Selection Algorithms to Select Energy-Efficient Servers for Storage and Computation Processes

It is now critical to reduce electric energy consumed by servers in a cluster, especially scalable systems like cloud computing systems. In a cluster of servers, most application processes like Web applications use not only CPU but also storages like HDD. In our previous studies, the power consumption MLPCS model and computation MLCMS model are proposed. Ways to energy-efficiently perform processes which use either CPU or storages are discussed in a cluster of servers. In this paper, we consider a more general type of process which does both the computation and accesses to storages. By using the MLPCMS and MLCMS models, we propose LEAS and GEAS algorithms to select servers to perform processes in a cluster so that the total electric energy consumption of the servers can be reduced. We evaluate the algorithms in terms of total electric energy consumption of the servers and average execution time of the processes. We show the electric energy consumed by servers can be reduced in the LEAS and GEAS algorithms.

Energy-Aware Algorithms to Select Servers in Scalable Clusters

It is critical to reduce the electric energy consumed in information systems, especially server clusters. In this paper, we discuss an MLPCM (multi-level power consumption with multiple CPUs) model and an MLCM (multi-level computation with multiple CPUs) model of a server with multiple CPUs. In this paper, we newly propose a modified globally energy-aware (MEA) algorithm to select a server for a process in a cluster of m servers. In the MEA algorithm, a server where a process all is to be performed is selected with computation complexity O(m) if the total electric energy of the servers is minimum. We evaluate the MEA algorithm and show not only the total electric energy consumption of the servers but also the average execution time of processes are reduced in the MEA algorithm compared with other algorithms.

Algorithms to Energy-Efficiently Select a Server for a General Process in a Scalable Cluster

Application processes like Web applications use not only CPU but also storages like HDD. In our previous studies, the algorithms to select a server in a cluster are proposed to energy efficiently perform processes which use either CPU or storages. In this paper, we consider a more general type of processes which do both the computation and access to storages. In this paper, we propose a computation model of a server to perform general processes and a pair of LEAG (locally energy-aware) and GEAG (globally energy-aware) algorithms to select servers to perform general processes in a cluster so that the total electric energy consumption of the servers can be reduced. We evaluate the LEAG and GEAG algorithms and show the electric energy consumed by servers and the average execution time of processes can be reduced in the LEAG and GEAG algorithms.

Simple Energy-Aware Algorithms to Selecting a Server for Storage and Computation Processes in a Cluster

In our previous studies, we propose a pair of LEAG (locally energy-aware) and GEAG (globally energy-aware) algorithms to select servers to perform processes which use CPU and storages in a cluster. However, it takes time to estimate the termination time of each current process on each server. In this paper, we newly propose a simple estimation model which uses only the number of processes performed on each server. Then, we propose a simple LEAG (SLEAG) and GEAG (SGEAG) algorithms to select a server. In the evaluation, we show the electric energy consumed by servers and the average execution time of processes can be more reduced in the SGEAG algorithm than the other algorithms.

Design and Evaluation of Algorithms to Energy-Efficiently Select Servers for Storage and Computation Processes

It is now critical to reduce electric energy consumed in a cluster of servers, especially scalable systems like cloud computing systems. In clusters, most application processes like web applications use not only CPU but also storages like databases. In this paper, we consider storage processes which read and write data in files in addition to computation processes. We propose an MLPCMS (power consumption model for a storage server) model which shows how much electric power a server consumes to perform storage and computation processes. We also propose an MLCMS (computation model for a storage server) model which shows the expected execution time of storage and computation processes. By using the MLPCMS and MLCMS models, we propose a GEAS (globally energy-aware server with storage processes) algorithm to select servers to perform computation and storage processes in a cluster so that the total electric energy consumption of the servers can be reduced. We evaluate the GEAS algorithm in terms of total electric energy consumption of the servers. We show the electric energy consumed by servers can be reduced in the GEAS algorithm.

Power Consumption and Computation Models of a Storage Server

It is now critical to reduce electric energy consumed in a cluster of servers, especially scalable systems including a huge number of servers like cluster computing systems. Types of application processes like computation, storage, and communication processes are performed on servers in clusters. In clusters, most applications use not only CPU resources but also storage drives like database and web applications. In this paper, we consider storage processes which read and write files in storage devices. The SPCS model (simple power consumption model for a storage server) shows how much electric power a server consumes to perform storage and computation processes. In our macro-level approach, we first measure the electric power consumed by a whole server to perform storage and computation processes and the computation time of each process. Then, we define the SPCS model of a server to perform storage and computation processes by abstracting parameters like number of processes which dominate the electric power consumption. We also define a simple computation model for a storage server (SPCS model) to perform storage and computation processes.