Abstract: Biological data contains very important information for genoma analysis. In last decades, the size of these data are constantly growing. So the Next Generation Sequence (NGS) data has been introduced. These kind of data are represented by different data format, such as FASTQ, FASTA, SAM, etc. In order to allow a good analysis and storing of them, due to large dimension of these data, several compressor have been performed. FQSqueezer is a novel genomic compressor for FASTQ data files. But several issues are present due to multithread version that runs on multi-core hardware. As known as the number of core in a CPU is limited and very minor with respect to GPUs’ cores number. In order to increase the performance related to this compressor method, in this work we present a GPU-parallel implementation of cited compressor by exploiting CUDA framework. More precisely, a suitable domain decomposition is able to give an appreciable gain of performance in time terms and reliability. Several execution tests confirm the gain of efficiency achieved by our parallel implementation.