Uploaded by Tejas Gowda

Outlay

advertisement
Name
1. Mohammed Rushad
2. Akshat Nambiar
Registration Number Roll Number Mobile Number Email
181437
181329
181CO232
181CO204
918105058850
919972012444
mohammedrushad@gmail.com
akshat1804@gmail.com
TOPIC: Green Software Engineering
TITLE: Optimization of a Big Data Compression algorithm for Energy Efficiency
ABSTRACT:
Big data has changed the way we use technology, but as the amount of data processed for any task keeps growing
exponentially, the large energy requirements cannot be ignored. Hence, large sets of data are often compressed using
established compression algorithms to reduce the amount of data to be stored or transmitted. But compression itself is
an energy intensive task, and given how often we are required to compress/decompress data, it is only reasonable to try
and optimize this process for reduced energy consumption. If one compressed file is copied to several locations, then it
must be decompressed multiple times at each destination. We are attempting to optimize these stages to reduce energy
consumption while processing Big Data.
OBJECTIVES:
1. To optimize existing ZStandard data compression software for energy efficiency during decompression as well as
compression.
2. To test improvements and changes on a variety of large sets of data to ensure correctness.
WORK METHODOLOGY:
1. Analyze existing software workflow and attempt to identify points of weakness that needs to be optimized. (Mid
September)
2. Implement the methods used in previous works for the ZStandard software. (End of September)
3. Design our solution to the problem by building off of previous work. (Mid October)
4. Implement the proposed design in the software. (End of October)
5. Test the implementation against a variety of data to ensure correctness, make improvements as required and
make final report documenting the work in the required format. (End of November)
OUTCOME: An energy efficient and improved ZStandard compression algorithm for processing large amounts of data
and able to handle various forms of data.
REFERENCES:
1. Joseph Azar, Abdallah Makhoul, Mahmoud Barhamgi, Raphaël Couturier: “An energy efficient IoT data
compression approach for edge machine learning”
2. Chi Yang a, Xuyun Zhang, Changmin Zhong, Chang Liu, Jian Pei, Kotagiri Ramamohanarao, Jinjun Chen: “A
spatiotemporal compression based approach for efficient big data processing on Cloud”
3. Jie Songa, Zhongyi Maa, Richard Thomas, Ge Yuc: “Energy efficiency optimization in big data processing platform
by improving resources utilization”
4. Issam Raïs, Daniel Balouek-Thomert, Anne-Cécile Orgerie, Laurent Lefèvre, Manish Parashar: “Leveraging
energy-efficient non-lossy compression for data-intensive applications”
5. Kenneth Barr and Krste Asanovic´: “Energy Aware Lossless Data Compression”
Download