HammingMesh: A Network Topology for Large-Scale Deep Learning
HammingMesh is optimized specifically for ML workloads and their communication patterns.
Torsten Hoefler
Advertisement
Author Archives
Disentangling Hype from Practicality: On Realistically Achieving Quantum Advantage
What are the promising applications to realize quantum advantage?
Enabling Highly Scalable Remote Memory Access Programming with MPI-3 One Sided
In this work, we design and develop bufferless protocols that demonstrate how to implement the MPI-3 RMA interface and support scaling to millions of cores.
ETH's Torsten Hoefler walks through a week of workshops, tutorials, papers, and his reactions to a week at SC.
Locality, Locality, Locality: A Note from SC14
Steadily growing chips, systems, and application parallelism are driving the movement of data across larger distances in systems, and data locality is becoming one of the major topics in efficient parallel computing. Locality has emerged as a significant theme at SC14, and research in this area is growing as researchers struggle to effectively manage this growing challenge to HPC application performance.
Building the HPC Community, Shaping Early Careers
The SC conference series provides a valuable opportunity each year to shape early research careers in an event where the whole community meets to exchange news, build collaborations, define research directions, and learn about technology.
Shape the Future of Computing
ACM encourages its members to take a direct hand in shaping the future of the association. There are more ways than ever to get involved.
Get Involved