Search arXiv⌕ Search
Home/Archive/arXiv

arXiv · 2607.26503

Safety-Gated Autoscaling: A Multi-Layered Defense Architecture for Kubernetes Vertical Resource Optimization

Azra Karakaya·Erva Şengül·Ahmet Kaplan
View original source ↗Read source document ↗Export metadata ↓

Abstract

Kubernetes is the standard platform for orchestrating containerized applications, yet resource management remains difficult. To stay safe, engineers over-provision CPU and memory, leaving reserved but unused capacity that is the main source of wasted cost. The built-in Horizontal and Vertical Pod Autoscalers are reactive: they act only after a threshold is crossed, which causes lag, over-provisioning, and can mask software defects by granting a leaking workload more memory. Predictive autoscalers focus on improving forecasting accuracy or run inside proprietary infrastructure, and anomaly detection is used only to alert, never to block a harmful action. The Intelligent Cluster Optimizer is an open-source Kubernetes operator that right-sizes container workloads with safety as a first-class concern. Its central contribution is a five-layer safety pipeline where a memory-leak detector, based on linear regression with R^2 scoring, acts as a blocking gate: if a leak is detected the recommendation is rejected, so the optimizer never hides a bug by enlarging a broken container. The pipeline combines SLA monitoring, a circuit breaker, HPA/PDB conflict detection, and a policy engine, with rollback and dry-run mode for human approval. Recommendations are produced by percentile analysis and Holt-Winters forecasting, balanced through multi-objective Pareto optimization at the per-container level. We validated the system with 1118 automated tests at 80.3% coverage and a live deployment on Google Kubernetes Engine, where right-sizing produced estimated cost savings of 20--40% in what-if projections and the leak gate reached 83% detection accuracy.

Explore related subjects

cs.DC

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Azra Karakaya, Erva Şengül, Ahmet Kaplan. 2026-07-29. Safety-Gated Autoscaling: A Multi-Layered Defense Architecture for Kubernetes Vertical Resource Optimization. https://arxiv.org/abs/2607.26503

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

arXiv2026

Concertina: Data-Centric Adaptive Pipeline Parallelism for Efficient Heterogeneous Long-Context LLM Training

Long context training is crucial for extending LLM context windows. Existing schemes, such as sequence parallelism, incur substantial communication overhead. Pipeline parallelism (PP) reduces this cost, but its effectiveness hinges on partitioning granularity. Batch-level PP employing sequence packing exhibits high memory consumption in long-context scenarios, whereas token-level PP splitting sequences into slices alleviates memory overhead but may introduce performance degradation. Moreover, the skewed sequence-length distribution in real-world datasets defeats any monolithic, static choice of PP granularity. In this paper, we propose \textit{Dynamic Pipeline Parallelism} (DPP), which transforms PP granularity from a static design choice into a workload-adaptive optimization space over packed, split, and hybrid chunks. DPP further introduces a new coupling between heterogeneous pipeline scheduling and gradient checkpointing. To solve this coupling, \name co-optimizes dynamic chunk scheduling with \textit{Stage-Aware Chunk-Level Adaptive Checkpointing}. Comprehensive experiments demonstrate that \name achieves up to 1.69\texttimes\ speedup over FlexSP and up to 1.40\texttimes\ over MEPipe. The source code is available at https://github.com/wsjdsg/InfiniPipe-code.

cs.DC↗
arXiv2026

A Low-Latency Semantic State Estimator using Latent Predictive Learning for Dynamic Network Monitoring and Orchestration

Closed-loop network monitoring and orchestration increasingly require semantic interpretations of live telemetry beyond raw counter collection. However, dynamic cloud-edge environments change both the active node set and the monitoring query at runtime, while control loops demand bounded millisecond-scale responses. We introduce a latent predictive state estimator (LPSE) for dynamic network monitoring and orchestration, built on latent predictive learning over streaming telemetry. The framework converts variable-cardinality node telemetry into topology-adaptive temporal representations, fuses them with monitoring questions, and returns bounded answers from a semantic codebook instead of autoregressive text generation. This design enables fixed-cost, single-pass inference while preserving semantic interpretability. By operating on permutation-invariant, slot-routed node representations keyed by stable identity, the model maintains a fixed input space and generalizes to node addition, removal, and reordering without retraining. Experimental results on a multi-node Kubernetes cluster show semantic prediction accuracy of 87.13% at approximately 72x lower mean inference latency compared to the lowest latency LLM tested.

cs.DC↗
arXiv2026

X-Stage: Modeling Post-Issue Backpressure in GPU Communication--Computation Fusion

Fine-grained, device-initiated communication allows fused GPU kernels to issue remote stores directly from their compute pipelines, a pattern increasingly used in expert parallelism (EP), tensor parallelism (TP), and Ulysses-style sequence parallelism (UP). Existing designs reason about where communication is issued and when remote data becomes ready, but lack a quantitative model of the sender-side interval after a remote store is accepted and before it becomes visible at the destination. This interval determines whether communication remains decoupled from computation or backpressures it. We identify X-Stage, a software-visible post-issue stage with finite decoupling. Downstream pressure can dissipate while the issuer resumes useful work, whereas sustained injection consumes X-Stage headroom and eventually stalls the compute pipeline. We characterize this behavior and build a calibrated model that predicts whether remote-store arrivals accumulate backpressure or recover during intervening computation. Guided by the model, we reshape bursty arrivals when they would exhaust X-Stage headroom and exploit natural compute windows when headroom can recover concurrently. Evaluation across representative EP, TP, and UP workloads shows up to 1.62x fused-kernel, 1.75x end-to-end, and 1.43x sender-visible speedup, respectively. Microbenchmarks further validate the model's predictions of backlog accumulation, recovery, and sender-side backpressure.

cs.DC↗