Lessons Learned on the Path to Guaranteeing the Error Bound in Lossy Quantizers
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , |
|---|---|
| Format: | Preprint |
| Publié: |
2024
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866909263570403328 |
|---|---|
| author | Fallin, Alex Burtscher, Martin |
| author_facet | Fallin, Alex Burtscher, Martin |
| contents | Rapidly increasing data sizes in scientific computing are the driving force behind the need for lossy compression. The main drawback of lossy data compression is the introduction of error. This paper explains why many error-bounded compressors occasionally violate the error bound and presents the solutions we use in LC, a CPU/GPU compatible lossy compression framework, to guarantee the error bound for all supported types of quantizers. We show that our solutions maintain high compression ratios and cause no appreciable change in throughput. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2407_15037 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Lessons Learned on the Path to Guaranteeing the Error Bound in Lossy Quantizers Fallin, Alex Burtscher, Martin Distributed, Parallel, and Cluster Computing Rapidly increasing data sizes in scientific computing are the driving force behind the need for lossy compression. The main drawback of lossy data compression is the introduction of error. This paper explains why many error-bounded compressors occasionally violate the error bound and presents the solutions we use in LC, a CPU/GPU compatible lossy compression framework, to guarantee the error bound for all supported types of quantizers. We show that our solutions maintain high compression ratios and cause no appreciable change in throughput. |
| title | Lessons Learned on the Path to Guaranteeing the Error Bound in Lossy Quantizers |
| topic | Distributed, Parallel, and Cluster Computing |
| url | https://arxiv.org/abs/2407.15037 |