Multi-tenancy is essential for unleashing SmartNIC's potential in
datace...
Cloud computing represents an appealing opportunity for cost-effective
d...
Numerous microarchitectural optimizations unlocked tremendous processing...
High-performance clusters and datacenters pose increasingly demanding
re...
Triangle count and local clustering coefficient are two core metrics for...
The allreduce operation is one of the most commonly used communication
r...
The growing size of data center and HPC networks pose unprecedented
requ...
Simple graph algorithms such as PageRank have been the target of numerou...
The capacity of offloading data and control tasks to the network is beco...
The interconnect is one of the most critical components in large scale
c...
The recent line of research into topology design focuses on lowering net...
Deep learning at scale is dominated by communication time. Distributing
...
System noise can negatively impact the performance of HPC systems, and t...
Applications often communicate data that is non-contiguous in the send- ...
Load imbalance pervasively exists in distributed deep learning training
...
We introduce FatPaths: a simple, generic, and robust routing architectur...
Nowadays simulations can produce petabytes of data to be stored in paral...