Matsuki Yoshino

APL as a prototyping language: case study of a compiler development project

We are applying prototyping method using APL to a commercial compilers development Project. This paper will discuss the following matters based on our one year experience of the project:<list><item>Merits of APL as a prototyping language. </item><item>An environment of the prototype. </item><item>A representation of tables and the intermediate language of compilers in an APL environment. </item><item>A strategy of transforming the APL prototype into final product written in Pascal. </item><item>Evaluation of this method. </item></list>

Classification of energy-saving operations from the perspective of system management

System management operations consist of detecting events that have occurred in a monitored system, checking system conditions, judging appropriate actions to be taken, and implementing the chosen actions. There are several patterns of system management operations. In this paper, the authors classify energy-saving operations in data centers according to patterns of system management operations. This classification helps enterprises consider what kind of energy-saving operations are feasible in current system management environments and what kind of improvement should be made in order to implement desired energy-saving operations. Use cases are provided for each pattern of system management operations. A quantitative estimation of power consumption reduction and analysis for improvement based upon the estimation are provided for one use case.

An anomaly detection method for individual service on web-based system by selection of dummy variables in multiple regression

This paper addresses detecting anomalies of individual services from their total resource usage on web-based system. Because the total resource usage is a linear combination of the number of accesses to each service, multiple regression analysis can be applied to estimate a resource usage per an access to each service as regression coefficient. However, the regression coefficients differ from the resource usage per an access of the services, which is caused by unstable resource usage per an access. We propose a method based on a multiple correlation coefficient to identify anomaly time and anomaly services. The proposed method identifies anomaly time when the correlation coefficient is decreased. And the proposed method identifies the anomaly service by judging whether the correlation coefficient is increased or not after the selection of the dummy variable. The experimental result shows that the proposed method can identify all the anomaly time, and improves precision rate and recall rate of detecting anomaly services by 20% at least, respectively.

Message analysis method based on a stream database for information system management

The authors conducted a detailed survey of a large-scale data center. They found that automated monitoring of messages produced by operating systems, middleware, and applications was adopted as a key method for detecting system failures and malfunctions, and found that this was an effective method for improving the quality of system management operations. The survey revealed that message occurrences are not only monitored, they are also analyzed for particular patterns. Those patterns include multiple occurrences surpassing a threshold in a predefined time period, consecutive occurrences within a predefined time interval, or occurrences violating predefined sequences, and so on. A custom-made application is currently used for this message analysis. The authors found some problems in implementation of the application. To overcome these problems and aid information system management, the authors proposed a message analysis method using a stream database. An experiment proved the effectiveness of the method.