Electronics and Telecommunications Research Institute

Thin Multimedia, Inc.

Samsung Electronics Co., Ltd.

Jin-Hwan Jeong

Yong-Ju Lee

Hag-Young Kim

Choon-Seo Park

Chang-Soo Kim

Song-woo Sok

Ok-Gee Min

Hyuck Yoo

Yoo-Hyun Park

Myung-Joon Kim

Tae-kwon Na

Won-Jae Lee

Jung-Keun Kim

Disclosed herein are an apparatus and method for predicting performance attributable to the parallelization of hardware acceleration devices. The apparatus includes a setting unit, an operation unit, and a prediction unit. The setting unit divides the time it takes to perform a task into a plurality of task stages and processing stages, and sets one of a parallelization index and target performance. The operation unit calculates the times it takes to perform the stages, and calculates at least one of the ratio of a target parallelization stage in the task and a speed improvement value. The prediction unit calculates an expected performance value or a parallelization index based on at least one of the calculated the times it takes to perform the stages, the calculated ratio of the target parallelization stage, the calculated speed improvement value, and the set target performance.

Apparatus and method for predicting performance attributable to parallelization of hardware acceleration devices

Provided are a method for generating an instantaneous decoding refresh (IDR) unit for trick play, and a trick play system and method using the same. The trick play system includes: a speed information input unit configured to receive a speed value; a transmission condition determining unit configured to determine instantaneous decoding refresh (IDR) transmission conditions based on the speed value inputted through the speed information input unit; a double-speed IDR-unit generating unit configured to generate a double-speed IDR-unit by adjusting the number and data size of video frames to be transmitted in accordance with the transmission conditions determined by the transmission condition determining unit; and a transmitting unit configured to transmit the generated double-speed IDR-unit to an external user terminal.

Method for generating double-speed IDR-unit for trick play, and trick play system and method using the same

Provided are a large-scale cluster monitoring system and a method for automatically building/restoring the same, which can automatically build a large-scale monitoring system and can automatically build a monitoring environment when a failure occurs in nodes. The large-scale cluster monitoring system includes a CM server, a BD server, GM nodes, NA nodes, and a DB agent. The CM server manages nodes in a large-scale cluster system. The DB server stores monitoring information that is state information of nodes in groups. The GM nodes respectively collect the monitoring information that is the state information of the nodes in the corresponding groups to store the collected monitoring information in the DB server. The NA nodes access the CM server to obtain GM node information and respectively collect the state information of the nodes in the corresponding groups to transfer the collected state information to the corresponding GM nodes. The DB agent monitors the monitoring information of the nodes in the groups, which is stored in the DB server, to detect a possible node failure.

Large-scale cluster monitoring system, and method of automatically building/restoring the same

Provided is an availability prediction method for a high availability. The method includes calculating a basic survival probability that the other node survives until a failure on one node of two nodes constituting a cluster is fixed, and determining an optimal number of nodes meeting a preset reference availability probability by calculating an availability probability for a predetermined range of the number of nodes on the basis of the basic survival probability. The method determines the number of nodes in the high availability cluster so as to match a reference availability probability, and is able to accomplish an optimal configuration of a cluster by calculating the availability probabilities for combinations between active node and passive nodes and between head nodes and switches.

Availability prediction method for high availability cluster

A removably connected power supply for supplying external power to an electronic product. An output unit supplies power to the electronic product. A switching unit selects whether to supply the power to the output unit, and a switching apparatus turns on the switching unit in response to being connected with the electronic product and turns off the switching unit in response to being disconnected from the electronic product. The switching apparatus includes a switch which performs a switching operation in response to the connection or disconnection with the electronic product from the output unit. A switching element controls operation of the switching unit in response to the operation of the switch. Thus, the power supply is operated in a power saving mode and unnecessary power consumption and damage caused by voltage or current spikes are prevented.

Power supply having power saving mode

There is provided an intermediate data based video/audio streaming method in which, in the transmission of video/audio data, a server transmits the video/audio data in the form of intermediate data and a client receives and directly processes it without using an additional decompression apparatus, allowing easy video/audio data transfer and simplifying the structure of the client system. The method includes: a video/audio transmitting step of, in case that the server transmits a video/audio file to the client, converting compressed video/audio data to be transmitted into intermediate data and duplicating the intermediate data into a packet to transmit it to the client; and, in case that the client processes the intermediate data transmitted through the network, an intermediate data receiving step of analyzing the received packet, when the analyzed result corresponds to a video intermediate data packet, reconstructing the original video data from the video intermediate data duplicated into the packet and then outputting it, and, when the received packet corresponds to an audio intermediate data packet, converting the audio intermediate data duplicated into the packet into analog voice data and outputting it.

Intermediate data based video/audio streaming method

There is provided an apparatus and method for streaming MPEG-1 data in which, when the MPEG-1 data is transmitted through the network, the MPEG-1 data is converted into intermediate data, a server transfers the intermediate data in the form of packet with synchronizing video intermediate data and audio intermediate data with each other, thereby allowing a client to be able to receive and process the intermediate data with a simple conversion. The apparatus includes: an MPEG-1 system stream decoder for decoding the system stream of an MPEG-1 file; a YUV data generator for converting a videos stream transmitted from the MPEG-1 system stream decoder into YUV data as intermediate data; a PCM data generator for converting an audio stream transmitted from the MPEG-1 system stream decoder into PCM data; and an MPEG-1 intermediate data server for synchronizing the YUV data and PCM data, respectively generated by the YUV data generator and PCM data generator, with a YUV PTS and PCM PTS obtained from the MPEG-1 system stream decoder and a system clock, to transmit them to the client.

Growing community of inventors

Seoul, South Korea

Jin-Hwan Jeong