DMA-Latte (MICRO'26) GPU communication, SDMA/Copy engine, Intra-server
UCCL-EP (OSDI'26) MoE, GPU communication, RDMA
WaveCC (SIGCOMM'26) CBFC, Wafer Scale, Molecular Dynamics, Co-design
FabricPerf (SIGCOMM'26) GPU communication, Profiling, NVLink
UCCL (OSDI'26) GPU communication, RDMA, Inter-server, Scale-out, Transport
MemChannel (NSDI'26) CXL pooling, Sender-driven congestion control
RamRyder (OSDI'26) CXL pooling, bandwidth provisioning, SW
PhaseWeave (ISCA'26) Heterogeneous chiplet architecture
Packet Transactions (SIGCOMM'16) Programmable switch
TiNA (ASPLOS'26) Host network, Chiplet, Sub-NUMA Clustering, DMA
LithOS (SOSP'25) GPU scheduling, Multi-tenancy
Other papers (keep updated)