Scanning HTML at Tens of Gigabytes per Second on ARM Processors
Fuente:
arXiv
Saved in:
| Main Author: | |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866910985290252288 |
|---|---|
| author | Lemire, Daniel |
| author_facet | Lemire, Daniel |
| contents | Modern processors have instructions to process 16 bytes or more at once. These instructions are called SIMD, for single instruction, multiple data. Recent advances have leveraged SIMD instructions to accelerate parsing of common Internet formats such as JSON and base64. During HTML parsing, they quickly identify specific characters with a strategy called vectorized classification. We review their techniques and compare them with a faster alternative. We measure a 20-fold performance improvement in HTML scanning compared to traditional methods on recent ARM processors. Our findings highlight the potential of SIMD-based algorithms for optimizing Web browser performance. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2503_01662 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Scanning HTML at Tens of Gigabytes per Second on ARM Processors Lemire, Daniel Data Structures and Algorithms Hardware Architecture Modern processors have instructions to process 16 bytes or more at once. These instructions are called SIMD, for single instruction, multiple data. Recent advances have leveraged SIMD instructions to accelerate parsing of common Internet formats such as JSON and base64. During HTML parsing, they quickly identify specific characters with a strategy called vectorized classification. We review their techniques and compare them with a faster alternative. We measure a 20-fold performance improvement in HTML scanning compared to traditional methods on recent ARM processors. Our findings highlight the potential of SIMD-based algorithms for optimizing Web browser performance. |
| title | Scanning HTML at Tens of Gigabytes per Second on ARM Processors |
| topic | Data Structures and Algorithms Hardware Architecture |
| url | https://arxiv.org/abs/2503.01662 |