G10: Enabling An Efficient Unified GPU Memory and Storage Architecture with Smart Tensor Migrations.
Haoyang ZhangYirui Eric ZhouYuqi XueYiqi LiuJian HuangPublished in: MICRO (2023)
Keyphrases
- storage devices
- memory management
- real time
- memory capacity
- data transfer
- higher order
- memory access
- heterogeneous computing
- random access
- secondary storage
- management system
- memory requirements
- associative memory
- data storage
- processing elements
- flash memory
- parallel implementation
- disk access
- file system
- compute intensive
- single instruction multiple data
- memory hierarchy
- memory space
- storage media
- disk storage
- multithreading
- computational power
- main memory
- high order
- neural network
- storage management
- limited memory
- parallel computation
- parallel computing
- shared memory
- operating system
- file management
- data structure