Unverified50% confidenceFactExact time
GPU的SM能同时驻留的Warp数量决定了GPU隐藏内存访问延迟的能力，制约Occupancy的核心资源是寄存器文件与共享内存
1
Sources
50%
Confidence
Long-term
Relevance
7/7/2026
First Seen
Sources
Related Claims
UnverifiedThe GPU memory hierarchy from fastest to slowest is: registers, L1 cache/shared memory, L2 cache, global memory (VRAM).79% similarUnverifiedGPU上下文切换需保存完整GPU状态（包括寄存器文件、共享内存和L1缓存），数据量可达数百MB，而CPU上下文切换仅需保存几KB77% similarUnverifiedCache expiration is essentially a trade-off providers make between GPU memory pressure and user experience, as maintaining large KV Caches over a long period consumes inference capacity.74% similarUnverified当专用GPU显存不够用时，Windows系统会自动将部分数据转移到从系统RAM划拨的共享GPU内存中73% similarVerified统一内存架构使CPU、GPU与NPU共享同一块内存池，避免数据在处理单元间反复搬运的开销73% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/187942API
curl https://kongchang.com/api/v1/knowledge/claims/187942MCP
get_claim(id=187942)