SLO-Aware Graph Forecasting for Intelligent Autoscaling in API and Microservice Backends

SLO-Scaler: Uncertainty-Aware SLO-Driven Autoscaling for Microservices

Autoscaling microservice-based applications to satisfy Service Level Objectives (SLOs) remains challenging due to bursty workloads, cascading latency across service dependencies, and cold-start overhead. Existing approaches such as the Kubernetes Horizontal Pod Autoscaler (HPA) rely on threshold-based CPU or memory metrics, which react too slowly to traffic spikes. Recent predictive methods improve responsiveness but generate point forecasts that ignore prediction uncertainty, leading to over-provisioning or oscillatory scaling. We propose SLO-Scaler, an uncertainty-aware autoscaling framework that predicts short-horizon request rates, tail latency, and SLO violation probability using a Bayesian LSTM model. SLO-Scaler integrates confidence-interval-based scaling decisions with a dependency graph analysis module that localizes bottleneck services, avoiding unnecessary whole-chain scaling. We evaluate SLO-Scaler on the DeathStarBench Social Network benchmark deployed on Kubernetes under periodic, bursty, and long-tail traffic patterns. Under bursty traffic, SLO-Scaler reduces the SLO violation rate by 29-56%, lowers the average replica count by 18-33%, and decreases scaling event frequency by 38-59% compared with the baselines, while achieving lower tail latency.

A Real-Time Workload Monitoring–Based Intelligent Auto-Scaling Framework for Cloud Systems

Kubernetes Horizontal Pod Autoscaler(HPA) and other existing auto-scaling solutions that respond reactively to demand experience significant delays in provisioning and inefficiencies when responding to sudden workload spikes. This paper proposes a new Real-Time Workload Monitoring-Based Intelligent Auto-Scaling Framework that uses a hybrid time-series forecasting model to proactively manage cloud computing resources. Specifically, it combines periodicity detection with Prophet-based models and long-term dependence capture via Long Short-Term Memory (LSTM) networks to provide advanced notice of CPU and memory requirements before load surges occur. The results of this approach were compared against traditional reactive auto-scaling methods and demonstrated that it significantly reduced scaling latencies, minimized service level agreement (SLA) violations, and optimized resource utilization during load surges.

CAPA: A PREDICTION-DRIVEN AUTOSCALING

AgileSphere: Predictive-Reactive Kubernetes Cluster Autoscaling With Warm-Pool Node Promotion

Container orchestration platforms have made cloud-native deployment routine, but autoscaling at the cluster level remains a persistent source of capacity-pressure service-level objective (SLO) violations under bursty workloads. Existing autoscalers either react slowly to capacity exhaustion, paying cloud cold-start latency on each scale-up, or scale only at the container level, leaving pending pods on saturated nodes. We present AgileSphere, a predictive-reactive cluster autoscaling framework that combines (i) ScalePlanner, an iTransformer-based capacity forecaster trained on the Alibaba 2022 cluster trace; (ii) ScaleBit, a magnitude-aware reactive controller that promotes pre-cordoned reserve nodes to active capacity in seconds—a warm-cache <inline-formula> <tex-math notation="LaTeX">$p_{95}$ </tex-math></inline-formula> of 4.30 s, rising to ~14 s only on a first cold image pull (vs. 60–180 s for cloud provisioning); and (iii) AgileServer, a monotone-max reconciler that arbitrates between the two desired-state sources. We implement the prototype on a live Kubernetes testbed and evaluate it under five conditions, including a Kubernetes Cluster Autoscaler (KCA)-equivalent port with modeled cloud cold-start latency. On a sustained-pressure workload across <inline-formula> <tex-math notation="LaTeX">$N=10$ </tex-math></inline-formula> seeds, AgileSphere reduces mean pending-pod-seconds (PPS) by <inline-formula> <tex-math notation="LaTeX">$23.3\times $ </tex-math></inline-formula> versus vanilla Kubernetes and <inline-formula> <tex-math notation="LaTeX">$2.2\times $ </tex-math></inline-formula> versus KCA at near-identical active node-seconds. A static-10-node over-provisioning ablation rules out the trivial alternative: AgileSphere delivers 80% lower backlog at comparable active node-seconds (within 5%) to a pre-uncordoned 10-node cluster, so the gain is not from “more capacity” but from “capacity at the right time.” Beyond mean improvement, the predictive arm contributes a <inline-formula> <tex-math notation="LaTeX">$5.2\times $ </tex-math></inline-formula> tighter seed-to-seed standard deviation than pure-reactive scaling, identifying a stability advantage that prior cluster-autoscaler work has not characterized. On an exploratory transient-pressure bursty workload (<inline-formula> <tex-math notation="LaTeX">$N=3$ </tex-math></inline-formula> seeds), AgileSphere remains statistically indistinguishable from the best reactive-only baseline on mean PPS (within 17%) while improving over KCA on both backlog (<inline-formula> <tex-math notation="LaTeX">$8.6\times $ </tex-math></inline-formula> lower) and active node-seconds (4% fewer) on the same workload. The reserve-pool promotion primitive achieves a warm-cache <inline-formula> <tex-math notation="LaTeX">$p_{95}$ </tex-math></inline-formula> of 4.30 s, approximately <inline-formula> <tex-math notation="LaTeX">$20\times $ </tex-math></inline-formula> tighter than modeled cloud cold-start latency; we also quantify the infrastructure cost of holding such a warm pool, which raises total provisioned capacity even where active node-seconds are comparable. These results show that hybrid predictive-reactive autoscaling delivers lower sustained-pressure backlog and higher operational stability than single-paradigm baselines, while remaining competitive with the best reactive-only configuration on transient workloads.

Zero-Instrumentation Dependency Discovery for Guided Microservice Migration Using eBPF

Migrating microservices across virtual machines (VMs) without knowledge of their runtime communication patterns risks creating cross-VM hotspots and latency spikes that are difficult to predict from static analysis alone. We use extended Berkeley Packet Filter (eBPF) kernel-level network tracing to automatically discover inter-service dependencies at runtime, with no application instrumentation, and use the resulting dependency graph to produce a traffic-aware migration plan ranked by return on investment (ROI). A two-pass process-identifier (PID) to port correlation algorithm recovers the identity of all 20 services in a shared-runtime testbed where processes are otherwise indistinguishable, matching the known ground-truth topology. The system discovers 32 dependency edges from 13,615 network events captured in three minutes, and applies spectral graph clustering with Kernighan-Lin refinement to partition services into VM-coherent groups. In simulation over the discovered graph, our ROI-ranked migration order reduces cumulative cross-VM traffic exposure during the migration window by 27% relative to alphabetical ordering, a deterministic proxy for arbitrary dependency-blind ordering. Collection overhead is mixed: in a controlled A/B test at near-saturation load on a host with two virtual CPUs (vCPUs), throughput fell by only 4.4%, but median (p50) latency rose by 383% and 99th-percentile (p99) latency by 1,050%. We therefore recommend running captures off-peak or on dedicated sampling nodes rather than under production saturation. All results are from a single 20-service testbed that we authored; we make no claim about behavior on production dependency graphs.

PCGAT-Stab: An Explainable Temporal Graph Attention Network for PCIe Protocol Behavior Modeling and System Stability Prediction in Cloud-Native Microservices

Cloud-native microservice architectures built on Kubernetes increasingly support AI SaaS platforms, financial-grade systems, and large-scale data centers, where high availability and low latency are critical requirements. However, while observability frameworks provide rich application- and orchestration-level metrics, they largely overlook the impact of low-level hardware behaviors, particularly PCIe protocol anomalies, on service stability. In modern GPU- and accelerator-intensive clusters, PCIe errors such as Completion Timeout, Replay Timer Expiration, and Link Retrain events can propagate across device topology and manifest as pod restarts, latency jitter, and SLA violations. To address this cross-layer visibility gap, we propose PCGAT-Stab, an explainable Temporal Graph Attention Network for PCIe protocol behavior modeling and system stability prediction in cloud-native microservices. The model constructs a dynamic PCIe topology graph, encodes protocol error propagation through multi-head graph attention, captures temporal accumulation effects via recurrent units, and maps hardware-level representations to pod-level risk embeddings using an explicit device–pod association mechanism. Furthermore, attention analysis and SHAP-based attribution provide interpretable insights into hardware-to-service causal relationships. Experiments on a real-world Kubernetes cluster dataset show that PCGAT-Stab achieves an AUC of 0.94 and an F1-score of 0.91 for instability classification, outperforming LSTM and Transformer baselines. For latency prediction, the model attains an RMSE of 5.72 and MAE of 4.68, demonstrating superior cross-layer modeling capability. These results confirm the effectiveness of topology-aware temporal graph attention for reliable system stability prediction in cloud-native microservices.