Shigeru Kusakabe

Generating supportive hypotheses in introducing formal methods using a software processes improvement model

We are investigating the effective facilitation methodology for the introduction of formal methods into actual software developments. One of the important issues is how to define and develop an adequate development process with formal methods. In this paper, we discuss this issue from a view point of software process improvement using a standard development process model, CMMI-DEV, which is a compilation of best practices in software development. We expect using the standard process model as a reference can facilitate common understanding for the advantages of formal methods and the comparison between actual instances of software process with formal methods. This approach leads to the exploitation of knowledge and findings obtained in the successful cases. We generate supportive hypotheses in the effect of formal methods on software development processes by using the standard improvement model. Then, we weakly examine them with a combination of a small-scale process, which has a strong relation to the organization level process, and a model-oriented formal method, VDM.

Applying Eco-Threading Framework to Memory-Intensive Hadoop Applications

Hadoop is a software framework for processing large data sets on clusters of commodity hardware. We apply our framework, which enhances performance and efficiency of memory-intensive multi-threaded applications, to Hadoop applications. The framework consists of a kernel-level thread scheduler, an application programming interface (API) for the scheduler, and a controller for the behavior of the scheduler through the API. We exploit the affinity of sibling threads, which have the same parent process and share the context, so that we can effectively exploit memory hierarchy by reducing memory-related undesirable events such as cache misses. We monitors performance metrics and automatically adjusts the behavior of the scheduler through the API to try to maximize the effectiveness of the scheduler. According to our preliminary evaluation result, our framework is promising to reduce the energy consumption of memory intensive Hadoop applications.

Visualizing centrality of process area networks in CMMI-DEV

In order to nd a clue to effectively introducing and utilizing new technology in improving software development process, we analyze and visualize the in-degree centrality of the process area networks in terms of the related process areas in CMMI-DEV. By visualizing the results of in-degree centrality analysis, we can have a perspective of process improvement in using advanced technology.

Monitoring Hadoop by Using IEEE1888 in Implementing Energy-Aware Thread Scheduling

Hadoop is a software framework for writing applications that process huge amount of data in parallel typically on clusters of commodity computing nodes. In implementing our energy-aware scheduling framework on a Hadoop system consisting of commodity nodes, we use IEEE1888, an open network protocol, Ubiquitous Green Community Control Network Protocol. Our scheduling framework is originally developed for single-node commodity platforms and consists of a kernel level thread scheduler, API, a monitor and a controller for the behavior of the scheduler. In order to implement this framework on a Hadoop cluster, we monitor and control the slave nodes over the network by using IEEE1888. Our framework monitors performance metrics and adjusts the behavior of the scheduler through the API. The framework can effectively reduce energy consumption for memory-intensive workload on thread-enabled Hadoop.

Requirements Development of Energy Management System for a Unit in Smart Campus

Institutional research supports collection of evidence, data, and information to conduct analyses for decision making, planning and continuous improvement in educational institutes. Conceptually, it covers wide range of areas such as admissions, student services, career outcomes, and facilities. It seems to become covering wider areas including emerging areas such as sustainability. As in todays social infrastructure, software plays an important role also in institutional domain. Software seems to keep growing in its scope, size and complexity also in institutional research domain. However, many organizations have not invested in the capabilities necessary to effectively manage life-cycle process for such products and services. We argue that software engineering techniques, such as life-cycle process model and software product line engineering are important in developing systems effective in institutional research. In this paper, we discuss a trial case of requirements development in improving the energy management system in our campus, as an emerging area of institutional research, by using software engineering techniques under the goal of increasing the awareness and involvement of members for smart energy management.

TOWARDS MAXIMIZING THROUGHPUT FOR MULTITHREADED PROCESSES IN LINUX

CPU scheduler is a very important key concept in operating system which affects scheduling objectives and criteria. Choosing or modifying scheduling policy for running threads depends on predefined and specific objectives and criteria. In this paper, we investigate the effectiveness of thread weight readjustment scheduler (TWRS) for multithreaded processes in multitasking operating systems from the view point of throughput. TWRS is a proportional share CPU scheduler designed for scheduling multithreaded processes. In this work, we make change kernel performance significantly by modifying just few key parameters. We propose TWRS which preallocates certain amount of CPU time to each thread of the multi-threaded processes. The scheduler was implemented and evaluated under specific hardware and software environment. According to our evaluation results, our scheduler is promising to optimize some scheduling criteria.

Preliminary evaluation of eco-threading framework for commodity platform

We are developing a framework to enhance performance and efficiency of multi-thread applications. The framework consists of a kernel-level thread scheduler, an application programming interface (API) for the scheduler, and a helper thread to control the behavior of the scheduler through the API. We exploit the affinity of sibling threads, which have the same parent process and share the context, so that we can reduce the occurrences of avoidable events such as cache misses and enhance the efficiency of multi-thread programs. The API enables us to adjust the control parameter to make a balance between the affinity and other issues in scheduling such as fairness. The helper thread monitors performance counters and automatically adjusts the behavior of the scheduler through the API to try to maximize the effectiveness of the scheduler. According to the results of the preliminary evaluation, our framework is promising to increase the energy efficiency as well as to reduce the execution time.

Analyzing Key Process Areas in Process Improvement Model for Service Provider Organization, CMMI-SVC

We investigate the effective facilitation methodology for introducing advanced technology in providing superior services. In the literature of process improvement models, we have a claim that effective processes provide a vehicle for introducing and using new technology in a way that best meets the business objectives. However, process improvement models have their complexity and we need a clue to use them as our vehicle. In this paper, we analyze relationships among process areas in a process improvement model for service provider organizations, CMMI-SVC, by focusing on Related Process Areas components. We observe process area networks whose nodes are process areas connected with their Related Process Areas components and we evaluate in-degree and between ness centrality in the process area networks for each maturity level in order to analyze key process areas.

Thread Weight Readjustment Scheduler: A Proportional Share CPU Scheduler for Multithreaded Processes

In this paper, we present thread weight readjustment scheduler (TWRS), a proportional share CPU scheduler designed for multithreaded processes. We show that weight readjustment enables existing schedulers to significantly optimize some scheduling criteria. We present thread weight readjustment scheduler, a proportional-share scheduler that is designed explicitly for multithreaded processes. Choosing or modifying scheduling policy for running and runnable threads depends on predefined and specific objectives and criteria. In this work, we make change kernel performance significantly by modifying just few key parameters. We propose a thread weight readjustment scheduler (TWRS) which preallocates certain amount of CPU time to each thread of the multi-threaded processes. The scheduler was implemented and evaluated under specific hardware and software environment. According to our evaluation results, our scheduler is promising to optimize some scheduling criteria, turnaround time in this context.

DEVELOPING CORE SOFTWARE REQUIREMENTS OF ENERGY MANAGEMENT SYSTEM FOR SMART CAMPUS WITH ADVANCED SOFTWARE ENGINEERING

Higher educational institutions such as universities provide campus data and information to stakeholders such as students, faculty staff, administrators, media, and community members. They also analyze such data and information for decision making, planning and continuous improvement. They obtain necessary capabilities for collecting, providing and analyzing data and information by acquiring products and services from suppliers or by developing products and services by themselves. As in today’s social infrastructure, software plays an important role and software seems to keep growing in its scope, size and complexity in infrastructure of educational institutions. However, many organizations have not invested in the capabilities necessary to effectively manage life-cycle process for such software products and services. We argue that software engineering techniques, such as process improvement framework and software product line engineering are important in developing requirements and systems effective in continuous improvement. In this paper, we discuss a trial case to improve the energy management system in our campus, as an emerging area of institutional continuous improvement, by using software engineering techniques under the goal of increasing the awareness and involvement of members for smart energy management.