Related papers: OptiNIC: A Resilient and Tail-Optimal RDMA NIC for…
As distributed machine learning (ML) workloads scale to thousands of GPUs connected by ultra-high-speed inter-connects, tail latency in collective communication has emerged as a primary bottleneck. Prior RDMA designs, like RoCE, IRN, and…
Today's data centers consist of thousands of network-connected hosts, each with CPUs and accelerators such as GPUs and FPGAs. These hosts also contain network interface cards (NICs), operating at speeds of 100Gb/s or higher, that are used…
We present OptiReduce, a new collective-communication system for the cloud with bounded, predictable completion times for deep-learning jobs in the presence of varying computation (stragglers) and communication (congestion and gradient…
Distributed learning is widely used for training large models on large datasets by distributing parts of the model or dataset across multiple devices and aggregating the computed results for subsequent computations or parameter updates.…
We present Laminar, the first TCP stack that delivers ASIC-class performance and energy efficiency on programmable Reconfigurable Match-Action Table (RMT) pipelines, providing flexibility while retaining standard TCP semantics and POSIX…
Emerging artificial intelligence (AI) and machine learning (ML) workloads present new challenges of managing the collective communication used in distributed training across hundreds or even thousands of GPUs. This paper presents STrack, a…
Next generation high-performance RDMA-capable networks will require a fundamental rethinking of the design and architecture of modern distributed DBMSs. These systems are commonly designed and optimized under the assumption that the network…
This paper introduces OptiGridML, a machine learning framework for discrete topology optimization in power grids. The task involves selecting substation breaker configurations that maximize cross-region power exports, a problem typically…
AI training and inference impose sustained, fine-grain I/O that stresses host-mediated, TCP-based storage paths. Motivated by kernel-bypass networking and user-space storage stacks, we revisit POSIX-compatible object storage for GPU-centric…
To fulfill the low latency requirements of today's applications, deployment of RDMA in datacenters has become prevalent over the recent years. However, the in-order delivery requirement of RDMAs prevents them from leveraging powerful…
Data center networks need to provide low latency, especially at the tail, as demanded by many interactive applications. To improve tail latency, existing approaches require modifications to switch hardware and/or end-host operating systems,…
The rapid scaling of large language models (LLMs) exacerbates communication bottlenecks in AI data centers (AIDCs). To overcome this, optical circuit switches (OCS) are increasingly adopted for their superior bandwidth capacity and energy…
Fast-evolving machine learning (ML) workloads have increasing requirements for networking. However, host network transport on RDMA NICs is hard to evolve, causing problems for ML workloads. For example, single-path RDMA traffic is prone to…
As the gap between network and CPU speeds rapidly increases, the CPU-centric network stack proves inadequate due to excessive CPU and memory overhead. While hardware-offloaded network stacks alleviate these issues, they suffer from limited…
High-performance clusters and datacenters pose increasingly demanding requirements on storage systems. If these systems do not operate at scale, applications are doomed to become I/O bound and waste compute cycles. To accelerate the data…
In the realm of edge computing, the increasing demand for high Quality of Service (QoS), particularly in dynamic multimedia streaming applications (e.g., Augmented Reality/Virtual Reality and online gaming), has prompted the need for…
In this paper, we consider how to provide fast estimates of flow-level tail latency performance for very large scale data center networks. Network tail latency is often a crucial metric for cloud application performance that can be affected…
SmartNICs have recently emerged as an appealing device for accelerating distributed systems. However, there has not been a comprehensive characterization of SmartNICs, and existing designs typically only leverage a single communication path…
Optimal transmission switching (OTS) improves optimal power flow (OPF) by selectively opening transmission lines, but its mixed-integer formulation increases computational complexity, especially on large grids. To address this, we propose a…
Rail-optimized network fabrics have become the de facto datacenter scale-out fabric for large-scale ML training. However, the use of high-radix electrical switches to provide all-to-all connectivity in rails imposes massive power, cost, and…