Kumiko Maeda

Frugal signal control using low resolution web-camera and traffic flow estimation

Due to rapid urbanization, large cities in developing countries have problems with heavy traffic congestion. International aid is being provided to construct modern traffic signal infrastructure. But often such an infrastructure does not work well due to the high operating and maintenance costs and the limited knowledge of the local engineers. In this paper, we propose a frugal signal control framework that uses image analysis to estimate traffic flows. It requires only low-cost Web cameras to support a signal control strategy based on the current traffic volume. We can estimate the traffic volumes of the roads near the traffic signals from a few observed points and then adjust the signal control. Through numerical experiments, we confirmed that the proposed framework can reduce an average travel time 20.6% compared to a fixed-time signal control even though the Web cameras are located at 500 m away from intersections.

Automatic Resource Scheduling with Latency Hiding for Parallel Stencil Applications on GPGPU Clusters

Overlapping computations and communication is a key to accelerating stencil applications on parallel computers, especially for GPU clusters. However, such programming is a time-consuming part of the stencil application development. To address this problem, we developed an automatic code generation tool to produce a parallel stencil application with latency hiding automatically from its dataflow model. With this tool, users visually construct the workflows of stencil applications in a dataflow programming model. Our dataflow compiler determines a data decomposition policy for each application, and generates source code that overlaps the stencil computations and communication (MPI and PCIe). We demonstrate two types of overlapping models, a CPU-GPU hybrid execution model and a GPU-only model. We use a CFD benchmark computing 19-point 3D stencils to evaluate our scheduling performance, which results in 1.45 TFLOPS in single precision on a cluster with 64 Tesla C1060 GPUs.

A parallel programming framework orchestrating multiple languages and architectures

This paper presents a novel parallel programming framework that orchestrates multiple languages such as C, C++, and Fortran and multiple computational architectures such as x86, POWER, and NVIDIAs Fermi to enhance productivity of parallel stencil applications while supporting high performance. Unlike traditional parallel programming frameworks, our framework provides three unique features: (1) simple meta-level, visual programming to construct workflows of components written in traditional programming languages, (2) optimal component parallelization and resource scheduling with the stencil communication pattern resolution, and (3) automatic network code generation including MPI, sockets, memory copies, and pointer passing. We prototyped incompressible computational fluid dynamics applications and demonstrate the effectiveness of our approach by evaluating our framework.

An event-processing system alerting analytically to networked vehicles

There are demands for alerts that use the results of intensive data analysis, such as relationships among events generated by many vehicles over a large area. These alerts require data analysis on servers. However, when a server receives massive amounts of data from vehicles, the processing latency increases because of the communication delays between the vehicles and the servers and the increased workload for data analysis on the server. Therefore we need to develop latency-tolerant alert-generating systems with scalable performance. In this paper, we report on a high-speed event-processing system architecture that integrates event processing in the in-vehicle systems with event processing in the servers. The in-vehicle system analyzes the vehicles sensor data, detects events, and sends packets of the event information to the servers. The server has a stream-processing system, a pre-aggregation system, and a full-data-accumulation system. The stream-processing system receives the packets from the in-vehicle system. The pre-aggregation system creates and updates decision tables for the alerts repeatedly. The alerter of the stream-processing system generates alerts from the table. We implemented a prototype system to generate alerts about obstacles. We tested input data from actual vehicles and a traffic simulator and estimated the vehicles can receive the alerts within 1.2 sec even when the server receives massive data from 120,000 vehicles, which meets the performance requirement for our alert scenarios.