14589

Parallel Decompression of Seismic Data on GPU Using a Lifting Wavelet Algorithm

Jairo A. Castelar, Carlos A. Angulo, Carlos A. Fajardo
Universidad Industrial de Santander, Bucaramanga
The Twentieth Symposium on Signal Processing, Images and Computer Vision (STSIVA-2015), 2015

@article{castelar2015parallel,

   title={Parallel Decompression of Seismic Data on GPU Using a Lifting Wavelet Algorithm},

   author={Castelar, Jairo A and Angulo, Carlos A and Fajardo, Carlos A},

   year={2015}

}

Download Download (PDF)   View View   Source Source   

704

views

Subsurface images are widely used by the oil companies to find oil reservoirs. The construction of these images involves to collect and process a huge amount of seismic data. Generally, the oil companies use compression algorithms to reduce the storage and transmission costs. Currently, the compression process is developed on-site using CPU architectures, whereas the construction of the subsurface images is developed on GPU clusters. For this reason, the decompression process has to be developed on GPU architectures. So, fast and parallel decompression algorithms are required to be implemented on GPUs. We implemented an algorithm that performs the decompression of seismic traces on GPU. The algorithm is based on a 2D Lifting Wavelet Transform. The decompression algorithm was developed in CUDA 6.5 and implemented into a GeForce GTX660 GPU. This algorithm was tested using different data sets supplied by an oil company. Experimental results allowed us to establish how the compression ratio affects the performance of our algorithm. Additionally, we also show how the number of threads per block affects this performance.
VN:F [1.9.22_1171]
Rating: 5.0/5 (1 vote cast)
Parallel Decompression of Seismic Data on GPU Using a Lifting Wavelet Algorithm, 5.0 out of 5 based on 1 rating

* * *

* * *

TwitterAPIExchange Object
(
    [oauth_access_token:TwitterAPIExchange:private] => 301967669-yDz6MrfyJFFsH1DVvrw5Xb9phx2d0DSOFuLehBGh
    [oauth_access_token_secret:TwitterAPIExchange:private] => o29ji3VLVmB6jASMqY8G7QZDCrdFmoTvCDNNUlb7s
    [consumer_key:TwitterAPIExchange:private] => TdQb63pho0ak9VevwMWpEgXAE
    [consumer_secret:TwitterAPIExchange:private] => Uq4rWz7nUnH1y6ab6uQ9xMk0KLcDrmckneEMdlq6G5E0jlQCFx
    [postfields:TwitterAPIExchange:private] => 
    [getfield:TwitterAPIExchange:private] => ?cursor=-1&screen_name=hgpu&skip_status=true&include_user_entities=false
    [oauth:protected] => Array
        (
            [oauth_consumer_key] => TdQb63pho0ak9VevwMWpEgXAE
            [oauth_nonce] => 1481142211
            [oauth_signature_method] => HMAC-SHA1
            [oauth_token] => 301967669-yDz6MrfyJFFsH1DVvrw5Xb9phx2d0DSOFuLehBGh
            [oauth_timestamp] => 1481142211
            [oauth_version] => 1.0
            [cursor] => -1
            [screen_name] => hgpu
            [skip_status] => true
            [include_user_entities] => false
            [oauth_signature] => v+AYssQenQZsRu1ibzD+V3egXs0=
        )

    [url] => https://api.twitter.com/1.1/users/show.json
)
Follow us on Facebook
Follow us on Twitter

HGPU group

2080 peoples are following HGPU @twitter

HGPU group © 2010-2016 hgpu.org

All rights belong to the respective authors

Contact us: