Javascript must be enabled to continue!
Mapreduce dalam Layanan Transcoding
View through CrossRef
Penyediaan file video dengan bitrate bervariasi menjadi syarat utama bagi layanan Video On Demand yang menerapkan adaptive streaming. Hal tersebut dilakukan dengan transcoding yang menghasilkan video dengan multi-bitrate. Proses multi-bitrate transcoding membutuhkan waktu yang tidak singkat. Lama durasi waktunya sebanding dengan besarnya bitrate, frame size dan frame rate. Untuk mempersingkat durasi transcoding, dibuat sebuah prototipe layanan transcoding dengan menerapkan Mapreduce. Layanan transcoding dengan Mapreduce terdiri dari satu komputer master dan beberapa komputer worker yang terhubung dalam satu LAN. Dengan dikoordinir oleh komputer master, komputer-komputer worker mengerjakan proses Map dan Reduce. Di dalam proses Map dilakukan transcoding terhadap segmented videos. Di dalam proses Reduce dilakukan penggabungan segmen-segmen video yang telah di-transcode dengan parameter/key (bitrate, frame size, dan frame rate) yang sama menjadi satu video yang utuh. Prototipe layanan transcoding dibuat menggunakan Library FFMPEG untuk transcoding, SCP untuk transfer file, RPC untuk komunikasi antar komputer. Di dalam pengujian prototipe, jumlah komputer worker ditentukan sebanyak 7 buah. Kinerja layanan transcoding sangat memuaskan dengan rata-rata efektifitas transcoding sebesar 71,3% dibandingkan dengan transcoding menggunakan satu komputer. AbstractThe provision of video files with varying bitrates is the main requirement for Video On Demand services that implement adaptive streaming. Transcoding that produces multiple bitrates achieves that. Multi-bitrate transcoding can take a longer time. That duration is comparable with the large of the video's bitrates, frame sizes, and frame rates. A transcoding service prototype was created based on Mapreduce to shorten the duration. Transcoding service using Mapreduce consists of several computers connected to a LAN, one as master and the other as worker computers. The master coordinates the workers to do the Map and Reduce process. In the Map process, workers transcode the segmented videos. In the Reduce process, workers merge all transcoded video segments with the same key or parameters (bitrate, frame size, and frame rate) into a single video. The transcoding service prototype was created using the FFMPEG library for transcoding, SCP for file transfer, and RPC for communication between computers. In the testing stage, the number of workers is 7 computers. Service performance is very satisfactory, with average transcoding effectiveness of 71,3% compared to transcoding using a single computer.
Fakultas Ilmu Komputer Universitas Brawijaya
Title: Mapreduce dalam Layanan Transcoding
Description:
Penyediaan file video dengan bitrate bervariasi menjadi syarat utama bagi layanan Video On Demand yang menerapkan adaptive streaming.
Hal tersebut dilakukan dengan transcoding yang menghasilkan video dengan multi-bitrate.
Proses multi-bitrate transcoding membutuhkan waktu yang tidak singkat.
Lama durasi waktunya sebanding dengan besarnya bitrate, frame size dan frame rate.
Untuk mempersingkat durasi transcoding, dibuat sebuah prototipe layanan transcoding dengan menerapkan Mapreduce.
Layanan transcoding dengan Mapreduce terdiri dari satu komputer master dan beberapa komputer worker yang terhubung dalam satu LAN.
Dengan dikoordinir oleh komputer master, komputer-komputer worker mengerjakan proses Map dan Reduce.
Di dalam proses Map dilakukan transcoding terhadap segmented videos.
Di dalam proses Reduce dilakukan penggabungan segmen-segmen video yang telah di-transcode dengan parameter/key (bitrate, frame size, dan frame rate) yang sama menjadi satu video yang utuh.
Prototipe layanan transcoding dibuat menggunakan Library FFMPEG untuk transcoding, SCP untuk transfer file, RPC untuk komunikasi antar komputer.
Di dalam pengujian prototipe, jumlah komputer worker ditentukan sebanyak 7 buah.
Kinerja layanan transcoding sangat memuaskan dengan rata-rata efektifitas transcoding sebesar 71,3% dibandingkan dengan transcoding menggunakan satu komputer.
AbstractThe provision of video files with varying bitrates is the main requirement for Video On Demand services that implement adaptive streaming.
Transcoding that produces multiple bitrates achieves that.
Multi-bitrate transcoding can take a longer time.
That duration is comparable with the large of the video's bitrates, frame sizes, and frame rates.
A transcoding service prototype was created based on Mapreduce to shorten the duration.
Transcoding service using Mapreduce consists of several computers connected to a LAN, one as master and the other as worker computers.
The master coordinates the workers to do the Map and Reduce process.
In the Map process, workers transcode the segmented videos.
In the Reduce process, workers merge all transcoded video segments with the same key or parameters (bitrate, frame size, and frame rate) into a single video.
The transcoding service prototype was created using the FFMPEG library for transcoding, SCP for file transfer, and RPC for communication between computers.
In the testing stage, the number of workers is 7 computers.
Service performance is very satisfactory, with average transcoding effectiveness of 71,3% compared to transcoding using a single computer.
.
Related Results
Mapreduce dalam Layanan Transcoding
Mapreduce dalam Layanan Transcoding
Penyediaan file video dengan bitrate bervariasi menjadi syarat utama bagi layanan Video On Demand yang menerapkan adaptive streaming. Hal tersebut dilakukan dengan transcoding yang...
Multi-constraint scheduling of MapReduce workloads
Multi-constraint scheduling of MapReduce workloads
In recent years there has been an extraordinary growth of large-scale data processing and related technologies in both, industry and academic communities. This trend is mostly driv...
High-Quality Live Video Streaming via Transcoding Time Prediction and Preset Selection
High-Quality Live Video Streaming via Transcoding Time Prediction and Preset Selection
Video streaming often requires transcoding content into different resolutions and bitrates to match the recipient's bandwidth and screen capabilities. Video encoders like H.264 off...
Optimizing data management for MapReduce applications on large-scale distributed infrastructures
Optimizing data management for MapReduce applications on large-scale distributed infrastructures
Optimisation de la gestion des données pour les applications MapReduce sur des infrastructures distribuées à grande échelle
Les applications data-intensive sont lar...
A Comparative Study of Join Algorithms in MapReduce
A Comparative Study of Join Algorithms in MapReduce
To analyze large volumes of data, a set of techniques is present in the IT community as the MapReduce paradigm, parallel RDBMS, column storage, and combinations of those technics. ...
Multimedia Transcoding in Mobile and Wireless Networks
Multimedia Transcoding in Mobile and Wireless Networks
Secure multimedia transcoding is a challenge that operates the encrypted multimedia content directly. For example, the encrypted multimedia data’s bit rate is changed directly in o...
Semantic-Based Video Transcoding Architectures for Quality of Service Applications in Mobile and Wireless Video Communication
Semantic-Based Video Transcoding Architectures for Quality of Service Applications in Mobile and Wireless Video Communication
Delivering streaming video over wireless networks is an important component for most interactive multimedia applications running on personal wireless handset devices. Such personal...
Extreme Rate Distributed Video Transcoding System
Extreme Rate Distributed Video Transcoding System
This chapter provides a comprehensive awareness and understanding of research efforts in the field of extreme rate-distributed video transcoding. The basic concepts and theories of...

