SearcharxivSearch

arXiv subjects

Mark Handley

Publications and source records attributed to Mark Handley.

6 recordsLinked to original sources

The Multipath Reliable Connection (MRC) Transport

MRC is an open, production-grade transport designed for large-scale AI/ML training over best-effort Ethernet. It extends RoCEv2 with explicit, composable primitives for per-packet multipath and sender-based congestion control, decouples packet delivery from semantic processing, adds multiple new capabilities for accelerated packet-loss recovery and adds resilience against port and path failures. This paper presents MRC and details its core capabilities and mechanisms.

cs.NI

Resilient AI Supercomputer Networking using MRC and SRv6

Tail latency dominates the performance of synchronous pretraining jobs when running at very large scales. We describe a three-pronged approach: (1) a new RDMA-based transport protocol, MRC, sprays across many paths and actively load-balances between them, eliminating the issue of flow collisions (2) the use of multi-plane Clos topologies to get the benefits of high switch radix and redundancy, allowing training clusters well over 100K GPUs to be built as two-tier topologies while increasing physical redundancy, and (3) the use of static source-routing using SRv6 to allow MRC the freedom to bypass failures by itself. We describe our experiences running MRC and static SRv6 routing in production in OpenAI and Microsoft's largest training clusters, where it has been used to train the latest frontier models. We demonstrate how MRC allows AI training jobs to ride out many network failures that previously would have interrupted training.

cs.NI

20 Years in Life of a Smart Building: A retrospective

Operating an intelligent smart building automation system in 2025 is met with many challenges: hardware failures, vendor obsolescence, evolving security threats and more. None of these have been comprehensibly addressed by the industrial building nor home automation industries, limiting feasibility of operating large, truly smart automation deployments. This paper introduces KaOS, a distributed control platform for constructing robust and evolvable smart building automation systems using affordable, off-the-shelf IoT hardware. Supporting control applications and distributed system operations by leveraging containerisation and managed resource access, KaOS seeks to achieve flexibility, security, and fault tolerance without sacrificing cost-effectiveness. Initial evaluation confirms the practical feasibility of our approach, highlighting its potential to sustainably maintain and incrementally evolve building control functionalities over extended timeframes.

cs.DC

Ultra Ethernet's Design Principles and Architectural Innovations

The recently released Ultra Ethernet (UE) 1.0 specification defines a transformative High-Performance Ethernet standard for future Artificial Intelligence (AI) and High-Performance Computing (HPC) systems. This paper, written by the specification's authors, provides a high-level overview of UE's design, offering crucial motivations and scientific context to understand its innovations. While UE introduces advancements across the entire Ethernet stack, its standout contribution is the novel Ultra Ethernet Transport (UET), a potentially fully hardware-accelerated protocol engineered for reliable, fast, and efficient communication in extreme-scale systems. Unlike InfiniBand, the last major standardization effort in high-performance networking over two decades ago, UE leverages the expansive Ethernet ecosystem and the 1,000x gains in computational efficiency per moved bit to deliver a new era of high-performance networking.

cs.NI

SMaRTT: Sender-based Marked Rapidly-adapting Trimmed & Timed Transport

With the rapid growth of artificial intelligence (AI) workloads in datacenters, the Ultra Ethernet Consortium (UEC) has defined a new high-performance transport layer to deliver the required performance at scale. A core component of this new standard is the Network Signal-based Congestion Control (NSCC) algorithm. This paper presents SMaRTT, the algorithm that forms the basis of the UEC NSCC specification. SMaRTT is a sender-based congestion control algorithm that systematically combines delay, Explicit Congestion Notification (ECN), and optional packet trimming into a cohesive state machine for fast, fair and precise window adjustments with seamless multipath support. At its core lies the novel QuickAdapt algorithm that accurately estimates and rapidly adapts to available capacity. Our evaluation shows that SMaRTT outperforms existing datacenter congestion control algorithms like Swift, RoCE, and MPRDMA by up to 50% and provides superior fairness, validating the design choices made in the UEC standard.

cs.NI

Reliable Low-Delay Routing In Space with Routing-Oblivious LEO Satellites

Large networks of Low Earth Orbit (LEO) satellites are being built using inter-satellite lasers. These networks promise to offer low-latency wide-area connectivity, but reliably routing such traffic is difficult, as satellites are very resource-constrained and paths change constantly. We present STARGLIDER, a new routing system where path computation is delegated to ground stations, while satellites are routing-oblivious and exchange no information at runtime. Yet, STARGLIDER satellites effectively support reliability primitives: they fast reroute packets over near-optimal paths when links fail, and validate that packets sent by potentially malicious ground stations follow reasonable paths.

cs.NI