Chang Oan Sung

Power consumption scheduling for peak load reduction in smart grid homes

This paper presents a design and evaluates the performance of a power consumption scheduler in smart grid homes, aiming at reducing the peak load in individual homes as well as in the system-wide power transmission network. Following the task model consist of actuation time, operation length, deadline, and a consumption profile, the scheduler copies or maps the profile according to the task type, which can be either preemptive or nonpreemptive. The proposed scheme expands the search space recursively to traverse all the feasible allocations for a task set. A pilot implementation of this scheduling method reduces the peak load by up to 23.1% for the given task set. The execution time greatly depends on the search space of a preemptive task, as its time complexity is estimated to be <i>O</i> (<i>M</i>^<i>N</i>_<i>np</i> · (<i>M</i>^<i>M/2</i>)^<i>N</i>_<i>p</i>), where <i>M, N</i>_<i>np</i>, and <i>N</i>_<i>p</i> are the number of time slots, preemptive tasks, and nonpreemptive tasks, respectively. However, it can not only be reduced almost to 2% but also made stable with a basic constraint processing mechanism which prunes a search branch when the partial peak value already exceeds the current best.

Towards reliable OSGi framework and applications

Upcoming ubiquitous computing systems are required to operate in dynamic, diverse, unverified, and unpredictable operating environment. The OSGi (Open Service Gateway initiative) framework employs the service-oriented approach and the java classloader architecture for the runtime service deployment, that are well suited to the dynamic environment envisioned for home networking and ubiquitous computing. However, the current OSGi framework does not provide full reliability measures, especially for failure conditions such as network, device, and application failures. This paper analyzes software reliability issues in OSGi framework and proposes a proxy-based reliable extensions. The design concept is implemented and partly tested on an open source OSGi platform, Oscar, for the smart home residential gateway test-bed.

Evaluation of interval-based dynamic voltage scaling algorithms on mobile Linux system

During the last several years, dynamic voltage scaling (DVS) algorithms are being used for energy consumption on real, fully functional battery supplied devices, adjusting the clock speed and supply voltage dynamically. Most DVS algorithms are investigated in interval-based and task-based strategies. Task-based algorithms consider task information, especially task deadline, on deciding what speed to choose at any given time. Interval-based algorithms predict the CPU speed of the upcoming interval based on observations of the CPU utilization of previous intervals, and then set the speed for that interval based on this prediction. Most DVS algorithms have only been tested in simulation environments. In this paper, those interval-based DVS algorithms are modified with different parameters on different workloads, and evaluated to know which one saves the most energy while not degrading computer performance.

AAGC: an efficient associativity-aware garbage collection scheme for hybrid FTLs

NAND ash memory has many advantages such as low-power consumption, shock-resistance, and non-volatility. For these reasons, NAND ash memory is used for data storage purposes in mobile, personal and even enterprise computer systems. When we use NAND ash memory as a means of data storage, it requires a garbage collection procedure due to its erase-before-write characteristic. The efficiency of this garbage collection is an important issue because it affects the performance of the NAND ash memory. In this paper, we propose an efficient garbage collection scheme that can be combined with previous hybrid FTLs. Our garbage collection scheme considers the associativity of the blocks when selecting a victim block. The proposed scheme effectively reduces the number of time-consuming block erasures. Our trace-driven simulation results show that the proposed garbage collection scheme can reduce the number of block erasures by as much as 14%. It can also reduce the average response time and maximum response time by nearly 10% and 70%, respectively.

Failure management development for integrated automotive safety-critical software systems

Nowadays the number of vehicles equipped with electronic components is increasing rapidly by replacing mechanical and hydraulic systems. The most advanced cars function appropriately via more than 50 Electronic Control Units (ECUs), sensors and actuators that exchange more than 2500 electronic signals among them. The electronic components are interconnected with automotive bus systems. There are several bus systems that have been developed or improved in automotive bus systems to meet the different requirements for automotive applications: Local Interconnection Network (LIN), Controller Area Network (CAN), FlexRay and Media Oriented System Transport (MOST). However, there are demands to combine these different bus systems to increase the efficiency and safety of the vehicle systems. Failure management is a most challenging problem in car industry since the integrated automotive system needs to communicate with software/hardware components on the different bus systems in a car. The discussion on how to interconnect those automotive bus systems in a safety-critical way is addressed in the paper, where failure management should be applied.

A group management scheme for an efficient location-based service

This paper proposes and evaluates the performance of a cooperative position fix scheme built on top of a group management mechanism supported typically in the embedded operating system. The main goal is to enhance the accuracy of map matching and thus the quality of diverse LBSs without an extra equipment. Designed on the telematics network consist of many in-vehicle telematics devices moving on the road network, the proposed scheme enables each device to collect the GPS readings from its neighbors through the air interface when necessary. With at least 3 affirmative points having a unique segment to match, the GPS deviation can be calculated from a system of indeterminate equations, and then used to improve positioning accuracy. The performance measurement result shows that the proposed scheme can reduce ambiguous points from 32.0 % to 20.6 %, while the number of points having affirmity larger than 3.0 was also increased from 61.0 % to 71.2 %, sharpening the gap between the best two matches.

ART: adaptive, reliable, and fault-tolerant task management for computational grids

The main goal of ART is reducing the number of replications by using checkpointing and rollback scheme for each replication. In ART, the minimum number of replications is adaptively selected based on analysis of probability of successful execution within the given deadline and reliability requirement of each task. Simulation results show that ART outperforms existing mechanisms.

An efficient patient monitoring system

The patient monitoring system depends on limited resources and needs to be configured using a variety of peripheral devices. Most of them have been used hardware-based BitBLT which provides better speed, but it is expensive. In this paper, we propose and implement a patient monitoring system to verify the performance of a software-based 2D BitBLT engine. To validate the proposed method, we tested a PCB evaluation board with an ARM9T core CPU. In the result, our proposed system shows reasonable performance even using software-based 2D graphics engine.

A Cooperative Position Fix Scheme Based on a Group Management on the Vehicular Network

This paper proposes and evaluates the performance of a cooperative position fix scheme built on top of a group management mechanism carried out by processes distributed across the vehicular network. The main goal is to enhance the accuracy of map matching and thus the quality of diverse location-based services without extra equipment. Designed on the telematics network consisting of many vehicles simultaneously moving on the road network, the proposed scheme enables the in-vehicle telematics device to play a role of collecting GPS readings from its neighbors through an appropriate wireless network interface when necessary. With at least 3 affirmative points having just a single segment to match, the GPS deviation for each time instant can be estimated by solving an equation system which has more variables than equations, and the estimated term improves the accuracy of positioning data. The performance measurement result shows that the proposed scheme can reduce the ratio of ambiguous points from 32.0% to 20.6%, while increasing that of affirmative points from 61.0% to 71.2%, sharpening the gap between the best two matches.