Amplifying Reliability with AWS Observability

Amplifying Reliability with AWS Observability Implementation Guide and best practices.

1. 24% of organizations have
breached a contractual service level
agreement in the last 12 months
66% of organizations use
between 2-5 monitoring or
observability tools.
Catchpoint SRE Report 2024
AWS User Group Colombo

2. Amplifying Reliability with
AWS
Observability Implementation
Guide and Best practices
Indika Wimalasuriya
AWS User Group Colombo
AWS User Group Colombo

3. Agenda •Why Observability Matters
•Observability vs. Monitoring
•AWS Observability Offerings
•AWS Observability Maturity Model
•AWS Tools & Services Overview
•Demo: Implementation Walkthrough
•Best Practices & Success Metrics
AWS User Group Colombo

4. Quick Intro about myself
•Reliability Engineering Advocate, Solution Architect (specializing in SRE,
Observability, AIOps, & GenAI)
•Senior Systems Engineering Manager at Virtusa, overseeing technical
delivery, capability development and offering.
•Passionate Technical Trainer.
•Energetic Technical Blogger.
•AWS Community Builder - Cloud Operations.
•Ambassador at DevOps Institute (PeopleCert).
AWS User Group Colombo

5. Managing Ever-Growing Complexity in Distributed
Systems
Monolith Microservices
On Premises Cloud Serverless
Expansion of Data SourcesSurge in Data Volume Exponential Rise in Failure
Scenarios
AWS User Group Colombo

6. Reliability: The Backbone of Modern Technology
Why consistent performance matters more than ever.
AWS User Group Colombo

7. Monitoring vs Observability
AWS User Group Colombo
Monitoring shows what’s visible above the surface, like tracking known issues, while observability reveals the deeper,
hidden insights needed to understand system behavior
Monitoring - Tracks predefined metrics and alerts on
known issues.
Observability - Provides insights into a system's internal
state by analyzing unknown or complex patterns based
on logs, metrics, and traces

8. Understanding Observability
AWS User Group Colombo
•Exactly “how” can the internal state of a system be known?
•Examples of Signals
•Data emitted and collected from these signals
With proper applications in place, forms of communication called signals are emitted that provide quality
information to monitor the internal state of the system known as Observability
•Metrics
•Events
•Logs
•Traces
•Telemetry

9. Metrics, Events, Logs and Tracers (MELT)
AWS User Group Colombo
Metrics are the values pertaining to
a system/application at a certain
point in time
Events are specific sequences of
occurrences that take place within a
system being monitored
Logs are the original data type; in their
most fundamental form, logs are
essentially lines of text a system or
application produces when certain code
blocks are executed
Traces, or more precisely, “distributed
traces”, are samples of causal chains
of events or transactions between
different components in a
microservices ecosystem

10. Performance Impact Business Outcomes
Observability Enables Detecting Slowness
AWS User Group Colombo
“SLOW is the new DOWN”
Walmart found that for every 1 second improvement in page load time, conversions
increased by 2%
COOK increased conversions by 7% by reducing page load time by 0.85 seconds
Mobify found that each 100ms improvement in their homepage's load time resulted in a
1.11% increase in conversion

11. AWS User Group Colombo
The observability market is
valued at approximately $12
billion USD, making it a
highly competitive space
with numerous major
players

12. AWS User Group Colombo
AWS Observability Native Offerings
AWS CloudWatch
Digital Experience
Monitoring
Insights & Analytics
Visualizations
Foundations
Instrumentation & Collection
Synthetics RUM
Application Signals
Container InsightsLambda Insights Log Insights
Application InsightsEC2 Health Live Trail
Dashboards Metric Explore SLOs
Metrics Logs Tracers
CloudWatch Agent AWS Distro for OpenTelmetry

13. Not Having a Plan is the
Biggest Observability
Anti-Pattern!

14. AWS Observability Maturity Model
Journey through Observability implementation
AWS User Group Colombo
APM
Standardize
Alerts
Infrastructure
Monitoring
Availability
based alerts
RUM
Metric
Anomaly
Baseline driven
issue detection
and corelation
AI driven Self
Diagnostic
(GenAI)
Enable Metrics
Measure SLOs
Metric
Forecasting
Standardize
Logs
Observability
as Code
Service Map
XLA based
alerts Log Anomaly
Rule base
Resolution
Workflows
AI driven Self
Healing (
GenAI)
Synthetic
Monitoring
Topology
Noise
Reduction
Runtime Code
Performance
Monitored
(Keeping
Lights –on)
Observable
(Deeper
Insights)
Corelated
(Holistic
View)
Predictable
(Proactive
Monitoring)
Autonomous
(Intelligent
Automation)

15. Level 1 - Monitored
Keeping Lights-On
AWS User Group Colombo
Infrastructure Monitoring •CloudWatch
Synthetic Monitoring •CloudWatch Synthetics
Availability-Based Alerts •CloudWatch Alarms

16. Level 2 - Observable
Deeper Insights
AWS User Group Colombo
APM (Application Performance Monitoring)•X-Ray
Standardize Logs
•CloudWatch Logs
•AWS OpenSearch
Enable Metrics
•CloudWatch
•AWS Distro for OpenTelemetry
Runtime Code Performance •CodeGuru
Standardize Alerts •CloudWatch Alerts
Observability as Code
•CloudFormation
•Terraform

17. Level 3 - Corelated
Holistic View
AWS User Group Colombo
Real User Monitoring (RUM) •CloudWatch RUM
Service Map •X-Ray Service Maps
Unified Topology •X-Ray Service Maps
Measure SLOs •CloudWatch Dashboards
Enable Correlation
•X-Ray Service Maps
•DevOps Guru
XLA Based Alerts •CloudWatch, X-Ray

18. Level 4 - Predictable
Pre-emptive Monitoring
AWS User Group Colombo
Metric Anomaly Detection •CloudWatch Anomaly Detection
Log Anomaly Detection •CloudWatch Log Anomalies
Metric Forecasting •AWS Forecast
Noise Reduction •CloudWatch Events
Baseline-Driven Issue Detection •DevOps Guru
Rule-Based Resolution Workflows
•Lambda
•AWS Systems Manager

19. Level 5 - Autonomous
Intelligent Automation
AWS User Group Colombo
AI-Driven Self-Diagnosis
•Amazon Lookout for Metrics
•GenAI
AI-Driven Self-Healing
•GenAI
•AIOps workflows via Systems Manager and Lambda

20. Demo
AWS User Group Colombo

21. Best practices
Standardize Logging & Monitoring
Use CloudWatch Logs & Metrics; ensure consistent log formats.
Instrumentation with X-Ray
Implement distributed tracing with X-Ray for visibility.
Automated Alerting & Response
Set CloudWatch Alarms; automate responses with Lambda/SNS.
Continuous Performance Optimization
Use Compute Optimizer for resource analysis and recommendations.
Integration with Managed Services
Leverage RDS, DynamoDB, Lambda with built-in CloudWatch
monitoring.
AWS User Group Colombo

22. Measure Progress with Business Outcomes
Mean Time to Detect (MTTD)
Reduce issue identification time.
Mean Time to Resolve (MTTR)
Shorten time to fix issues.
Mean Time Between Failures (MTBF)
Increase time between system failures.
Improved Reliability & Availability
Boost uptime and minimize downtime.
Enhanced User Experience
Improve satisfaction with faster
interactions.
Optimized Resource Utilization
Use resources efficiently to save costs.
AWS User Group Colombo
Increased Development Velocity
Speed up feature delivery and updates.
Alignment with SLOs
Meet performance targets and business
goals.

23. Stay Connected for the Latest on AWS Observability,
SRE & AIOps
AWS User Group Colombo
Connect with me on LinkedIn
– Indika Wimalasuriya
https://www.linkedin.com/in/indika-
wimalasuriya/
Follow my insights on Dev.tohttps://dev.to/indika_wimalasuriya

24. Thank you.