Comparing Managed Memory and ATS with and without Prefetching on NVIDIA Volta GPUs.
Rahulkumar GayatriKevin GottJack DeslippePublished in: PMBS@SC (2019)
Keyphrases
- prefetching
- response time
- access patterns
- hit rate
- web prefetching
- graphics processing units
- graphics processors
- access latency
- computational power
- cache misses
- caching scheme
- web documents
- graphics hardware
- web caching
- gpu implementation
- user perceived latency
- web logs
- cache replacement
- web page prediction
- general purpose
- web objects
- hit ratio
- main memory
- memory requirements
- buffer pool
- parallel implementation
- memory management