2 papers
cs.AR2026
CXL-ClusterSim: Modeling CXL-based Disaggregated Memory Cluster for Pooling and Sharing using gem5 and SST
Kaustav Goswami, Maryam Babaie, Hoa Nguyen +2
Large-scale AI training and inference require hundreds of gigabytes to terabytes of DRAM with high peak to average utilization ratios, resulting in overprovisioning. In cloud compu…
cs.AR2024
TDRAM: Tag-enhanced DRAM for Efficient Caching
Maryam Babaie, Ayaz Akram, Wendy Elsasser +6
As SRAM-based caches are hitting a scaling wall, manufacturers are integrating DRAM-based caches into system designs to continue increasing cache sizes. While DRAM caches can impro…