Prometheus exporters: tutorial and best practices
Learn how Prometheus exporters expose third-party metrics, how to build a simple exporter, and what practices keep metrics useful.
KubernetesMonitoring
All blog articles
Practical guides to building, running, and improving ML and AI in production.
All articles
Page 15 of 21
Prometheus metrics: types, capabilities, and best practices
Understand Prometheus metric types, collection architecture, storage behavior, custom metrics, and Kubernetes monitoring use cases.
KubernetesMonitoring
How to use Kubernetes metrics
Find Kubernetes CPU, memory, object-state, and control plane metrics, and read them correctly with kubectl and your monitoring tools.
KubernetesGuides
Correct evaluation labels and rescore saved predictions
Keep corrected labels as a new dataset version, reuse predictions when model inputs are unchanged, and compare evaluation runs in Polyaxon.
DataOpsMLOpsLineage
Rotate workload TLS with Kubernetes Pod certificates
Use Pod certificate projections and trust bundles for workload TLS, with signer prerequisites and rotation handling for long-running ML services.
KubernetesSecurityReliability
Pull private images with Polyaxon workload service accounts
Attach a private-registry pull secret to a Kubernetes ServiceAccount, select it for a Polyaxon operation, and verify the new Pod can use the intended image.
ProductKubernetesSecurity
Build reproducible training snapshots from Kafka
Define a bounded Kafka record window, preserve its output and offset manifest, and run the snapshot job on Polyaxon.
DataOpsKafkaMLOps
Evaluate a SageMaker endpoint from Polyaxon
Call an existing SageMaker endpoint from a Polyaxon job and retain model-specific predictions and evaluation results.
MLOpsSagemakerEvaluation
Hand off Polyaxon model artifacts to KServe
Move an approved model artifact from a Polyaxon run to a separately managed KServe InferenceService.
MLOpsKServeModel Serving
Investigate Polyaxon workload health with Prometheus
Correlate Kubernetes object state, container usage, and node health when investigating a Polyaxon workload.
KubernetesPrometheusMonitoring
Monitor Polyaxon services with Prometheus
Instrument application request metrics in a Polyaxon service and configure a separate Prometheus collector to scrape them.
KubernetesPrometheusMonitoring
Optimize multi-cloud costs for ML workloads
Control multi-cloud ML spending with normalized allocation, workload unit economics, placement policies, elastic capacity, and reproducible efficiency measurements.
FinopsCost OptimizationMLOps
Run an existing TFX pipeline in a Polyaxon job
Wrap a local TensorFlow Extended pipeline in one Polyaxon job while preserving its own metadata and artifacts.
MLOpsTensorFlowPipelines
Serve a TensorFlow SavedModel with Polyaxon
Package a versioned SavedModel with TensorFlow Serving and run its REST endpoint as a Polyaxon service.
MLOpsTensorFlowModel Serving
Submit a Polyaxon workflow from AWS Lambda
Hand a trusted operation from a short AWS Lambda invocation to Polyaxon while accounting for retries and uncertain responses.
MLOpsAWSAutomation
Trigger Polyaxon workflows from Kafka events
Connect validated Kafka events to fixed Polyaxon operations with a durable submission ledger and replay handling.
MLOpsKafkaAutomation
Kubernetes monitoring for ML workloads
Monitor Kubernetes control planes, nodes, containers, schedulers, applications, and ML outcomes with useful correlations and controlled cardinality.
KubernetesMonitoringMLOps
Minikube for local Kubernetes development
How to use Minikube for local Kubernetes development, basic cluster operations, and testing ML workloads before moving to shared infrastructure.
KubernetesGuides
Kubernetes storage classes overview
Understand Kubernetes StorageClasses, dynamic provisioning, reclaim policies, volume binding, and persistent volume claims.
KubernetesGuides
Datadog vs. CloudWatch for AWS ML platforms
Choose between Datadog, Amazon CloudWatch, or a combined approach for EKS and ML workloads by testing coverage, ownership, portability, and cost.
AWSObservabilityMonitoring
How to deploy Postgres on Kubernetes
Deploy a single PostgreSQL instance with a Secret, persistent storage, and private access, then verify that its data survives a Pod replacement.
KubernetesGuides
Azure Storage best practices for ML data and artifacts
Design Azure Storage accounts for ML datasets and artifacts with clear boundaries, workload identities, network controls, lifecycle rules, and recovery tests.
AzureStorageMLOps
Kubectl command not found: causes and fixes
Diagnose kubectl command not found errors caused by missing installs, PATH issues, permissions, Docker state, or Minikube command syntax.
KubernetesGuides
How to use the Kubernetes control plane
Understand the Kubernetes control plane, its core components, and the operational practices that keep scheduling and reconciliation healthy.
KubernetesGuides