1887

Abstract

Summary

Mixed precision formats on CPU and GPU enable lossy compression of data with extremely fast access to compressed values. While the compression ratio is less than other state-of-the-art methods like ZFP or SZ, this fast access should yield a significant performance boost for memory bandwidth bound computations, e.g., stencil operations or matrix-vector multiplication. We investigate this approach for data in a seismic application.

Loading

Article metrics loading...

/content/papers/10.3997/2214-4609.2025643027
2025-10-06
2026-02-15
Loading full text...

Full text loading...

References

  1. Anzt, H., Griitzmacher, T. and Quintana-Orti, E.S. [2019] Toward a modular precision ecosystem for high-performance computing. International Journal of High Performance Computing Applications, 33(6), 1069–1078.
    [Google Scholar]
  2. Di, S. and Cappello, F. [2016] Fast Error-Bounded Lossy HPC Data Compression with SZ. In: 2016 IEEE International Parallel and Distributed Processing Symposium (IPDPS). 730–739.
    [Google Scholar]
  3. Kriemann, R. [2025] Hierarchical Low-Rank Arithmetic with Floating Point Compression. SIAM Journal on Scientific Computing, 47(4), B763–B784.
    [Google Scholar]
  4. Lindstrom, P. [2014] Fixed-Rate Compressed Floating-Point Arrays. IEEE Transactions on Visualization and Computer Graphics, 20(12), 2674–2683.
    [Google Scholar]
/content/papers/10.3997/2214-4609.2025643027
Loading
/content/papers/10.3997/2214-4609.2025643027
Loading

Data & Media loading...

This is a required field
Please enter a valid email address
Approval was a Success
Invalid data
An Error Occurred
Approval was partially successful, following selected items could not be processed due to error