NetDAM: Network Direct Attached Memory with Programmable In-Memory Computing ISA
Kevin Fang, David Peng

TL;DR
NetDAM introduces a programmable network-attached memory architecture with in-memory computing capabilities, enabling high-performance data transfer and processing for large-scale distributed applications, demonstrated through FPGA prototype and MPI-Allreduce case.
Contribution
It presents a novel FPGA-based architecture that combines network-attached memory with programmable in-memory computing for efficient data-intensive operations.
Findings
FPGA prototype demonstrates high performance in MPI-Allreduce
NetDAM offers a flexible alternative to RDMA for data transfer
Effective integration of memory pooling and in-network computing
Abstract
Data-intensive applications like distributed AI-training may require multi-terabytes memory capacity with multi-terabits bandwidth. We directly attach the memory to the ethernet controller with some programable logic to design an efficient hardware "template" for Memory pooling and in-memory / in-network computing. We built an FPGA prototype of the NetDAM, andwe demonstrate MPI-Allreduce communication case, the NetDAM can be used as a software and hardware friendly programmable architeture with high performance alternative for RDMA.
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsParallel Computing and Optimization Techniques · Advanced Memory and Neural Computing · Advanced Data Storage Technologies
