Yoshinari Nomura

Mint: Booting Multiple Linux Kernels on a Multicore Processor

In this paper, we describe the implementation and evaluation of the Mint operating system, which runs multiple Linux instances by partitioning cores, memory blocks, and I/O devices. The proposed method is not a type of virtualization, hence, it makes the most efficient use of available cores. By performing some evaluations, we confirmed that all OS instances in Mint have performances equal to those in the original Vanilla Linux. In addition, because all the instances are completely independent of one another, Mint will not be affected by performance issues, even under the forthcoming many-core environments. In addition, each instance is fully armored Linux. This means that users can still use existing software assets.

Run-time Updating of Network Device Drivers

Recent operating systems are capable of using dynamic loadable module functions to update the device driver programs without shutting down the entire system. However, even if a system continues to operate, some services provided by the device driver will be temporarily unavailable during the run-time update. To reduce the period during which the services are unavailable, we propose a new method for run-time updating of device drivers; this method uses context migration from an old driver to an updated one. In this study, we describe the implementation of a network-device driver using our method. In addition, we have shown some advantages of our method by performing a trial on the FreeBSD operating system.

OS Mechanism for Continuation-based Fine-grained Threads on Dedicated and Commodity Processors

Fine-grained multithreading based on a natural model, such as dataflow model, is promising in achieving high efficiency and high programming productivity. In this paper, we discuss operating system issues for fine-grained multithread programs. We are developing an operating system called CEFOS based on a dataflow based computation model. A program on CEFOS consists of zero-wait threads which run to completion without suspension once started. Firing control among such threads is performed in a dataflow manner along with continuation relations in the program. Target platforms include fuce processor, which is dedicated to fine-grained multithreading, and commodity processors such as Intel x86. In this paper, after introducing our basic model and our operating system model, we discuss implementation issues on fuce and commodity platforms. The evaluation results indicate that our approach on commodity platforms is effective in reducing overheads while our approach on a special architecture naturally exploit parallelism even in I/O handling.

Bring Your Own Calendar: A CalDAV-Based Virtual Calendar System

Today, the sharing of schedule information across individuals or organizations has been made simpler by the Internet and personal devices such as smartphones. However, it is not as easy in bring-your-own-device environments due to problems related to privacy and cultural gaps among organizations and individuals. Current calendar systems do not have any easy, fine-grained calendar sharing scheme to support such environments. In this paper, we propose a new calendar system that employs a fine-grained calendar sharing scheme called Virtual Calendar, which is based on our CalDAV proxy engine. Our calendar system satisfies such fine-grained demands as ``I want to share my calendar with my wife for only those articles that start after 5 pm or on weekends.

A Practical Calendaring System Conforming with Ambiguous Pattern of Recurring Tasks

In a typical office, there are many recurring tasks. If the recurrence pattern can be determined, then this pattern can be used to forecast which task might occur or provide guidance for a coworker who takes on the given responsibilities. A calendaring system is a popular tool for task enumeration. However, the existing major implementations are not capable of handling fluctuation cycle in the recurrences or the relationships between the tasks, and thus these systems do not give the user any analytical perspective on the tasks. Therefore, this paper proposes a new calendaring system that can handle changes in the recurrences and display the relationships between tasks. Three new concepts, Task, Mission, and Job are introduced to describe the relationships between the tasks. Interoperability with the existing systems is also described.

Scalability of continuation-based fine-grained multithreading in handling multiple I/O requests on FUCE

Multi-tasking operating systems not only handle concurrent tasksbut also have concurrency and parallelism of various granularities in themselves. We claim that operating systems need to be developed based on a computation model that can deal with concurrency and parallelism of various granularities. In order to investigate this claim, we have been developing an operating system called CEFOS based on a dataflow-based computation model. A program for CEFOS consists of zero-waitthreads, each of which runs to completion without suspension once started. Synchronization between zero-wait threads is performed in a dataflow manner according to their continuation relations. Handler routines for I/O devices are also realized with zero-wait threads and executed under the continuation-based multithreading mechanism. We can eliminate interrupts that interfere with the execution ofinstruction streams in typical conventional approaches, and we can naturally handle concurrency and exploit parallelism in programs even for I/O-centric computation. In this paper, after introducing our model and our operating system based on the model, we discuss implementation issues on Fuce, which is a continuation-based multithreading processor dedicated to fine-grained multithreading. Then we evaluate the scalability of our system with the number of execution units and I/O devices.

A Practical Method for Forecasting the Future Calendar Events of Ambiguous Recurrence

We proposed a model for detecting patterns in tasks that recur irregularly. This model may be useful for predicting future tasks, but it is not clear how to extract the necessary information from the model. This paper proposes a method for forecasting future tasks by extracting statistical information from the model. First, we introduce some additional terms for expressing the statistical information. Next, we propose a new method for forecasting. Finally, we provide typical examples of recurring tasks and evaluate the method using these examples.

Implementation of a Practical Calendaring System Conforming with Ambiguous Pattern of Recurring Tasks

We proposed a model that is capable of handling the patterns of ambiguous recurring tasks. If we can grasp the pattern, it would be useful for prediction of the tasks which might happen in the future, and also be the hint on handing them over to co-workers. To utilize our model, we have developed a concrete method for manipulation or visualization of the modeled data. This method is useful to build a nifty user interface to the calendaring systems. This paper shows an implementation of calendaring system conforming with ambiguous recurring tasks. First of all, we recall our model. Next, we propose some basic operations for the modeled data, EDIT, READ, and INHERIT. Finally, we describe the implementation of the method as an interface to our calendaring system.

Acquiring Application History for Successful Handover of Business Tasks

In a typical office environment, employees are expected to collect information about their tasks, in the event that they must hand them over to others. This task information is usually stored in multiple business applications. In this paper, we investigated how task-related information is stored across multiple applications through an actual workflow. In addition, we address two problems encountered when collecting work information from applications. First, it is difficult to collect information that directly relates to the work unit or task unit. Second, it is unclear what type of information should be handed over at the beginning of tasks. To solve these problems, we propose a new method for acquiring work history from multiple applications. The proposed method manages the application history associated with work unit or task unit. Additionally, the proposed method manages the priorities of histories. Finally, we prove that the proposed method can effectively collect work information.

Memory Segmentation and Transfer in Mint Operating System

In order to utilize computer resources efficiently, technologies have been developed to enable running multiple OSes on a single computer. In our study, we developed the Mint operating system, which enables running multiple Linux instances on a multi-core processor. Mint is different from traditional virtualization technologies in that Mint can achieve performance isolation between Linux instances by segmenting computer resources. Currently, Mint statically segments and assigns its physical memory to each Linux instance. By transferring physical memory ranges among the Linux instances depending on their physical memory utilization, Mint can adapt to the load variations. In this paper, we describe the requirements for segmentation and transfer of physical memory and provide a solution.