2 papers
cs.OS2026
RetainZ: Reclaim-Time-Aware Placement for AI Checkpoints on Zoned SSDs
Minxing Chu, Ruoxi Yang
Solid State Drives (SSDs) are becoming the dominant medium for performance-critical storage. Zoned Namespace (ZNS) SSDs are getting more and more attractive because sequential writ…
cs.AR2026
Heterogeneity-Aware Microscaling for Efficient Low-Bit LLM Inference
Junyi Luo, Xinting Jiang, Tai-Hao Wen +9
Microscaling (MX) is now the standard for low-bit large language model (LLM) inference. Its 4-bit form MXFP4 still loses substantial accuracy, because existing MX formats fix eithe…