Tracing Matters! by Ronak Banka

Two of the biggest challenges with operations is visibility and response, to know what’s wrong and know that quickly. Tracing of modern apps can not only help Developers to improve app performance, but also help cloud operation teams to support in an efficient way.

This talk will cover what to trace and how it helped a super app to improve user experience.

Transcript

1. Ronak Banka TRACING MATTERS! devopsdays.org/singapore | @devopsdaysSG
2. HOW TRACING WORKS & DEMO WHAT IS DISTRIBUTED TRACING? WHY DISTRIBUTED TRACING? 01 02 AGENDA 03
3. WHY DISTRIBUTED TRACING?
4. LACK OF VISIBILITY!
5. WHY HAPPENED? • Users are Scaling at a Rapid Speed. • Engineering Organizations are Scaling to Ship Business Value Faster • Monoliths are getting transformed to Microservices
6. MICROSERVICES ARCHITECTURE CAN GET VERY COMPLEX!
7. With More Complexity, comes more Failure Scenarios!
9. BUT We have Monitoring in Place!! WHY Do we even need Tracing based Observability?
10. Monitoring is about what we Already Know, Observability is about Answering questions!
11. • If a request failed, where did the errors happen? • If there is latency in the app, where are the bottlenecks? • When Processing a request, which all other services are involved and what they did? • Performance Behaviors • Root Cause Analysis in case of an Incident? Distributed Tracing can Answer
12. WHAT IS DISTRIBUTED TRACING?
13. Distributed tracing refers to methods of observing requests as they propagate through distributed systems It is also a diagnostic technique that reveals how a set of services coordinate to handle individual user requests.
14. • Trace represents an end-to-end request; made up of single or multiple spans • Span represents work done by a single-service with time intervals and associated metadata; the building blocks of a trace • Tags metadata to help contextualize a span CONCEPTS OF TRACING?
15. • Trace Identifier: trace_id is generated in the root span and is immutable. • Parent Identifier: span_id of the parent span that spawned the current operation HOW ARE TRACES CONNECTED? Source: https://blog.sentry.io/distributed-tracing-101-for-full-stack-
16. Various Tracing Tools
18. Confidential and proprietary. © Temus Pte Ltd 2022 22 Service Map
19. Confidential and proprietary. © Temus Pte Ltd 2022 24 Traces
20. Confidential and proprietary. © Temus Pte Ltd 2022 25 Metrics