All Integrations
CloudCloudWatch Metric Streams + IAM

AWS Neptune Integration

Monitor graph query performance, Gremlin/SPARQL latency, and storage utilization across your Neptune graph database clusters. Get AI-powered traversal spike detection and replication lag alerting.

Setup

How It Works

01

Create IAM Role for Metric Streams

Provision an IAM role with CloudWatch permissions scoped to the AWS/Neptune namespace. TigerOps uses this role to deliver Neptune cluster and instance metrics via Firehose.

02

Deploy CloudWatch Metric Streams

Run the TigerOps CloudFormation stack to stream the AWS/Neptune namespace. Gremlin and SPARQL query latency, HTTP requests, and replication metrics begin flowing immediately.

03

Enable Neptune Audit and Slow Query Logs

Configure Neptune to publish slow query and audit logs to CloudWatch Logs. TigerOps ingests these logs and correlates slow traversals with query volume spikes.

04

Configure Graph Query SLO Alerts

Set latency SLOs per query type (Gremlin vs SPARQL vs openCypher). TigerOps fires alerts when query latency deviates from baselines and correlates with graph size growth.

Capabilities

What You Get Out of the Box

Graph Query Performance

Gremlin, SPARQL, and openCypher query request rates, error rates, and latency per query language. Identify which query types are driving CPU and memory pressure.

Gremlin & SPARQL Latency

P50, P90, and P99 latency per query language with historical trending. TigerOps alerts when traversal latency increases correlate with graph vertex or edge count growth.

Storage Utilization

Volume used storage, volume read/write IOPS, and free local storage per instance. Track graph data growth rates and predict when storage capacity upgrades are needed.

Replication Lag Monitoring

Per-replica replication lag in milliseconds for Neptune read replicas. Alert before replica lag causes read queries to return stale graph traversal results.

CPU & Buffer Cache Metrics

CPU utilization, buffer cache hit ratio, and freeable memory per Neptune instance. Low buffer cache hit ratios indicate graph traversal patterns that exceed in-memory capacity.

AI Query Spike Correlation

TigerOps detects anomalous graph query rates and correlates spikes with application deployments or new feature rollouts that introduce expensive traversal patterns.

Configuration

CloudFormation Stack for Neptune Metric Streams

Deploy the TigerOps CloudFormation stack to stream Neptune graph database metrics in minutes.

tigerops-neptune-streams.yaml
# TigerOps CloudFormation — Neptune Metric Streams
# aws cloudformation deploy \
#   --template-file tigerops-neptune-streams.yaml \
#   --stack-name tigerops-neptune \
#   --capabilities CAPABILITY_IAM

Parameters:
  TigerOpsApiKey:
    Type: String
    NoEcho: true

Resources:
  TigerOpsNeptuneStream:
    Type: AWS::CloudWatch::MetricStream
    Properties:
      Name: tigerops-neptune-stream
      FirehoseArn: !GetAtt TigerOpsDeliveryStream.Arn
      RoleArn: !GetAtt MetricStreamRole.Arn
      OutputFormat: opentelemetry0.7
      IncludeFilters:
        - Namespace: AWS/Neptune

  TigerOpsDeliveryStream:
    Type: AWS::KinesisFirehose::DeliveryStream
    Properties:
      HttpEndpointDestinationConfiguration:
        EndpointConfiguration:
          Url: https://ingest.atatus.net/api/v1/cloudwatch
          AccessKey: !Ref TigerOpsApiKey
        RequestConfiguration:
          CommonAttributes:
            - AttributeName: service
              AttributeValue: neptune
        RetryOptions:
          DurationInSeconds: 60

# Enable Neptune slow query logging:
# aws neptune modify-db-cluster-parameter-group \
#   --db-cluster-parameter-group-name my-neptune-params \
#   --parameters ParameterName=neptune_enable_audit_log,ParameterValue=1 \
#   --parameters ParameterName=neptune_query_timeout,ParameterValue=120000

# Recommended alert thresholds:
# GremlinRequestsPerSec spike > 3x baseline → Warning
# ReplicationLag > 5000ms → Critical
# BufferCacheHitRatio < 90% → Warning
FAQ

Common Questions

Which Neptune metrics does TigerOps collect?

TigerOps collects all AWS/Neptune CloudWatch metrics including GremlinRequestsPerSec, SparqlRequestsPerSec, GremlinWebSocketOpenConnections, CPUUtilization, FreeableMemory, FreeLocalStorage, VolumeReadIOPs, VolumeWriteIOPs, ReplicationLag, BufferCacheHitRatio, and EngineUptime.

Does TigerOps support Neptune Serverless?

Yes. Neptune Serverless publishes metrics to the AWS/Neptune namespace with the same dimension set as provisioned clusters. TigerOps provides Neptune Capacity Units (NCU) utilization dashboards specific to serverless clusters alongside standard latency metrics.

Can TigerOps monitor Neptune global databases?

Yes. Neptune global database secondary clusters emit metrics from each region they run in. TigerOps aggregates metrics across primary and secondary regions with per-region filtering and cross-region replication lag comparison panels.

How does TigerOps handle Neptune load from graph bulk load operations?

TigerOps tracks Neptune loader HTTP request metrics and bulk load status via EventBridge events. When a bulk load causes significant CPU or I/O spikes, TigerOps annotates the metric timeline with the load operation start and completion events.

Does TigerOps support openCypher query monitoring?

Yes. Neptune openCypher endpoints publish request metrics to CloudWatch as part of the AWS/Neptune namespace. TigerOps displays openCypher request rates and latency alongside Gremlin and SPARQL metrics on the same Neptune dashboard.

Get Started

Stop Guessing Why Neptune Graph Traversals Are Slowing Down

Gremlin and SPARQL latency tracking, storage trend monitoring, and AI query spike correlation. Deploy in 5 minutes.