SVE2 match instructions speed up JSON parsing in simd on ARM
lemire · x · 2026-09-20
Daniel Lemire follows up on his April claim that ARM's SVE2 match instruction could be the fastest way to match characters on ARM. What was once a toy benchmark has now been tested against a real parser: ARM engineer Madhurendra Purbay submitted a pull request to the simd library integrating SVE2-based character matching.
- ARM chips use NEON SIMD; newer processors add SVE, and the latest generation supports SVE2. Apple has not adopted SVE, but SVE-capable processors are available in the cloud.
- The integration into simd validates that the idea survives contact with production-grade JSON parsing, with real performance implications for the ubiquitous JSON format.
More from Infra
- VTrain, a Vulkan-based resident trainer, fixes memory leak and offloads more work to GPU — Savantskie1 · 2026-09-20
- vLLM ships day-0 support for Qwen-Image-2.1 with cross-step prefix KV cache — Alibaba_Qwen · 2026-09-20
- Dev builds helmstudio, an MLX-first local launcher for open models on Apple Silicon — janishar · 2026-09-20
- vLLM-Omni ships KV reuse, FP8 and CUDA Graph optimizations with Qwen — vllm_project · 2026-09-20
- GLM 5.3 Flash in NVFP4 quantization gets a local ChatGPT-style setup — TheZachMueller · 2026-09-20
- You can't measure OSS model usage — estimate labs' compute or use inference-provider tokens — xeophon · 2026-09-20