Increasing Efficiency of Data Transfer Between Main Memory and Intel Xeon Phi Coprocessor or NVIDIA GPUS with Data Compression

Besedin, Konstantin Y.; Kostenetskiy, Pavel S.; Prikazchikov, Stepan O.

doi:10.1007/978-3-319-21909-7_32

Cited by 6 publications

(1 citation statement)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…While flourishing to achieve high data processing capabilities, GPUbased compressors target high versatility and a wide range of usage scenarios, such as in-memory compression [33], compression of MPI messages [34], and reducing CPU-GPU data transfer time [35]. On the one hand, the current cuZFP only allows it to use where high compression throughput is the priority, as its compression quality is low compared to cuSZ under the same compression ratio.…”

Section: Problem Statementmentioning

confidence: 99%

FZ-GPU: A Fast and High-Ratio Lossy Compressor for Scientific Computing Applications on GPUs

Zhang¹,

Tian²,

Di³

et al. 2023

Proceedings of the 32nd International Symposium on High-Performance Parallel and Distributed Computing

View full text Add to dashboard Cite

Today's large-scale scientific applications running on high-performance computing (HPC) systems generate vast data volumes. Thus, data compression is becoming a critical technique to mitigate the storage burden and data-movement cost. However, existing lossy compressors for scientific data cannot achieve a high compression ratio and throughput simultaneously, hindering their adoption in many applications requiring fast compression, such as in-memory compression. To this end, in this work, we develop a fast and highratio error-bounded lossy compressor on GPUs for scientific data (called FZ-GPU). Specifically, we first design a new compression pipeline that consists of fully parallelized quantization, bitshuffle, and our newly designed fast encoding. Then, we propose a series of deep architectural optimizations for each kernel in the pipeline to take full advantage of CUDA architectures. We propose a warplevel optimization to avoid data conflicts for bit-wise operations in bitshuffle, maximize shared memory utilization, and eliminate unnecessary data movements by fusing different compression kernels. Finally, we evaluate FZ-GPU on two NVIDIA GPUs (i.e., A100 and RTX A4000) using six representative scientific datasets from SDRBench. Results on the A100 GPU show that FZ-GPU achieves an average speedup of 4.2× over cuSZ and an average speedup of 37.0× over a multi-threaded CPU implementation of our algorithm under the same error bound. FZ-GPU also achieves an average speedup of 2.3× and an average compression ratio improvement of 2.0× over cuZFP under the same data distortion. CCS CONCEPTS• Theory of computation → Massively parallel algorithms; Data compression.

show abstract

Section: Problem Statementmentioning

confidence: 99%