Learning in Chaos: Efficient Autoscaling and Self-healing for Distributed Training at the Edge
Frequent node and link changes in edge AI clusters disrupt distributed training, while traditional checkpoint-based recovery and cloud-centric autoscaling are too slow for scale-out and ill-suited to chaotic and self-governed edge. This paper proposes Chaos, a resilient and scalable edge distributed training system with built-in self-healing and autoscaling. It speeds up scale-out by using multi-neighbor replication with fast shard scheduling, allowing a new node to pull the latest training state from nearby neighbors in parallel while balancing the traffic load between them. It also uses a cluster monitor to track resource and topology changes to assist scheduler decisions, and handles scaling events through peer negotiation protocols, enabling fully self-governed autoscaling without a central admin. Extensive experiments show that Chaos consistently achieves much lower scale-out delays than Pollux, EDL, and Autoscaling, and handles scale-in, connect-link, and disconnect-link events within 1 millisecond, making it smoother to handle node joins, exits, and failures. It also delivers the lowest idle time, showing superior resource use and scalability as the cluster grows.
Code (0)
등록된 구현이 없습니다.
Tasks
SchedulingSimilar Papers 제목 키워드 기반
On Evaluating Self-Adaptive and Self-Healing Systems using Chaos Engineering
With the growing adoption of self-adaptive systems in various domains, there is an increasing need for strategies to assess their correct behavior. In particular self-healing systems, which aim to provide resilience and …
CHESS: A Framework for Evaluation of Self-adaptive Systems based on Chaos Engineering
There is an increasing need to assess the correct behavior of self-adaptive and self-healing systems due to their adoption in critical and highly dynamic environments. However, there is a lack of systematic evaluation me…
Taming the Chaos: Coordinated Autoscaling for Heterogeneous and Disaggregated LLM Inference
Serving Large Language Models (LLMs) is a GPU-intensive task where traditional autoscalers fall short, particularly for modern Prefill-Decode (P/D) disaggregated architectures. This architectural shift, while powerful, i…
Bio-inspired Agentic Self-healing Framework for Resilient Distributed Computing Continuum Systems
Human biological systems sustain life through extraordinary resilience, continually detecting damage, orchestrating targeted responses, and restoring function through self-healing. Inspired by these capabilities, this pa…
Hamiltonian dynamics and distributed chaos in DNA
It is shown that distributed chaos, generated by Hamiltonian DNA dynamics with spontaneously broken time translational symmetry, imprints itself on the DNA sequence of Arabidopsis thaliana (a model plant for genetic sequ…