Choon Seo Park

GPU accelerated item-based collaborative filtering for big-data applications

Recommendation systems are a popular marketing strategy for online service providers. These systems predict a customers future preferences from the past behaviors of that customer and the other customers. Most of the popular online stores process millions of transactions per day; therefore, providing quick and quality recommendations using the large amount of data collected from past transactions can be challenging. Parallel processing power of GPUs can be used to accelerate the recommendation process. However, the amount of memory available on a GPU card is limited; thus, a number of passes may be required to completely process a large-scale dataset. This paper proposes two parallel, item-based recommendation algorithms implemented using the CUDA platform. Considering the high sparsity of the user-item data, we utilize two compression techniques to reduce the required number of passes and increase the speedup. The experimental results on synthetic and real-world datasets show that our algorithms outperform the respective CPU implementations and also the naïve GPU implementation which does not use compression.

Real-Time Data Flow Language Processing System for Handling Streams of Data

Apache Pig system generates MapReduce jobs by compiling program scripts written in Pig Latin to process large data sets in parallel on distributed computing nodes. There are inefficient features in Pig due to the limitation of the MapReduce, e.g., the MapReduce is used only for batch processing. As various smart devices are extensively utilized recently, streams of data are generated explosively and the need to process streams of data in real-time is required. In this paper, we propose a data flow language processing system, called LAMA-CEP, by generating DAG-based stream processing services to process unbounded streams of data in real-time continuously. We present a stream processing language, called Pig Latin Stream extended from Pig Latin. Programs written in Pig Latin Stream are translated into distributed stream processing jobs and then the jobs are executed on a highly scalable distributed stream processing system to process large streams of data in real-time.

The design of digital items server for processing multi DI clients on ubiquitous environments

To stream adapted contents to a variety of devices such as PDA, PC, HDTV and so on, on ubiquitous environments, digital items (DI) client access DI server to get information for adapted contents streaming and to store data related with streaming service. When a lot of DI client concurrently connect DI server, to increase the performance of the DI server in working process related with adapted streaming, we propose new technique that there are some main controller threads and each main controller thread has its own FIFO, which manages DI messages. Because each main controller thread has its own FIFO, by distributing DI messages to some main controller thread that is less busy, we make DI server keep good performance and condition

The design and implementation of asset management system on the NGIS

We have developed the next generation Internet server (NGIS) which has network-storage (NS) Cards for streaming service. The NGIS system provides assets of various types to support streaming service and has NS cards with a powerful network function. The NGIS system offers streaming service for assets of various types, such as MPEG 1 files, MPEG 2 files and MPEG 4 files. In this paper, we propose the method of installing assets, such as MPEG 1 files, MPEG 2 files and MPEG 4 files, to support to streaming service in the NGIS system. If the type of multimedia data is MPEG 1 or 2, we propose the method of installing assets, which makes fast forward (FF) files and fast rewind (FR) files in real-time in order to support a FF service and a FR service, and suggest another method which makes a FF file and a FR file off-line and installs the FF file and the FR file when system of installing assets begins. In this paper, also, we propose contents management system (CMS) calling asset installer automatically to install assets when the assets are at a certain place.