10 papers
Trivance: Latency-Optimal AllReduce by Shortcutting Multiport Networks
Anton Juerss, Vamsi Addanki, Stefan Schmid
AllReduce is a fundamental collective communication operation in distributed computing and a key performance bottleneck for large-scale training and inference. Its completion time…
Birkhoff Decompositions and Photonic Interconnects Wait! Don't Forget the Compute!
Eliezer Amponsah, Vamsi Addanki
The growing demand for efficient communication in distributed training and inference has sparked significant interest in reconfigurable photonic interconnects across both academia…
BShare: Packet Queueing Delay-Driven Buffer Sharing for Datacenter Switches
Krishna Agarwal, Muhamad Rizka Maulana, Vamsi Addanki +1
Modern datacenter switches share packet buffers across ports to boost overall throughput and reduce packet loss. However, as buffer availability per-port-per-bandwidth unit continu…
Harvest: Adaptive Photonic Switching Schedules for Collective Communication in Scale-up Domains
Mahir Rahman, Samuel Joseph, Nihar Kodkani +2
As chip-to-chip silicon photonics gain traction for their bandwidth and energy efficiency, their circuit-switched nature raises a fundamental question for collective communication:…
The Harmonic Policy for Online Buffer Sharing is (2 + ln n)-Competitive: A Simple Proof
Vamsi Addanki, Julien Dallot, Leon Kellerhals +2
The problem of online buffer sharing is expressed as follows. A switch with output ports receives a stream of incoming packets. When an incoming packet is accepted by the switc…
When Light Bends to the Collective Will: A Theory and Vision for Adaptive Photonic Scale-up Domains
Vamsi Addanki
As chip-to-chip silicon photonics gain traction for their bandwidth and energy efficiency, collective communication has emerged as a critical bottleneck in scale-up systems. Progra…