5 papers
CCAaaS: An Orchestration Platform for High-Velocity Design, Development, and Deployment of Data Center Transports at Meta
Prankur Gupta, Balasubramanian Madhavan, Miao Xu +7
Today, data center networks run multiple congestion control algorithms (CCAs), each tailored to distinct traffic characteristics, from latency-sensitive intra-region RPCs to high-t…
From ASIC to Fleet: Lessons from Building and Operating a Hyperscaler NIC
Prankur Gupta, Alexander Duyck, Jakub Kicinski +25
We describe the operational infrastructure built to deploy and operate fbnic, a custom multi-host NIC, across hundreds of thousands of production hosts at Meta. Vendor multi-host N…
HCCL: Collective Communication for Meta Training and Inference Accelerators
Wesley Bland, Tiago Antunes, Lars Paul Huse +63
We present HCCL, a collective communication library co-designed with Meta's MTIA 300 accelerator, the first Meta chip to integrate backend networking directly on chip package. MTIA…
Reasoning-based Anomaly Detection Framework: A Real-time, Scalable, and Automated Approach to Anomaly Detection Across Domains
Anupam Panwar, Himadri Pal, Jiali Chen +4
Detecting anomalies in large, distributed systems presents several challenges. The first challenge arises from the sheer volume of data that needs to be processed. Flagging anomali…
Bayesian Model Averaging for Data Driven Decision Making when Causality is Partially Known
Marios Papamichalis, Abhishek Ray, Ilias Bilionis +2
Probabilistic machine learning models are often insufficient to help with decisions on interventions because those models find correlations - not causal relationships. If observati…