Design patterns for cloud architecture

Slides

https://docs.google.com/presentation/d/1Lu36kP2j3dx-hyD9gOX6TMDuseF_j7SZSpCdDLy-wF8/edit?usp=sharing

Update as of 20/01/2023

Please see https://prathamesh-sonpatki.notion.site/The-hidden-cost-of-Instrumentation-f5e597ef2d444cc8b41f4226a9d9bf69 for a detailed brief of this talk.

Abstract

The path to Observability starts with instrumentation. If you are just starting the Observability journey of your organization, then setting up and maintaining tools and pipelines can be expensive and time-consuming.

There are many different ways of instrumenting applications and components that affect observability, such as log aggregation, distributed tracing, application performance management, and synthetic monitoring. Each method has its own pros and cons.

In order to determine the best way of instrumentation for the observability of software systems, it is important to consider the specific requirements of the system and the cost/benefit of each monitoring method. Different solutions may be more appropriate for different systems depending on the size, complexity, and budget of the organization.

Session outline

The talk will be divided into the following broad sections.

• The landscape of instrumentation in cloud-native applications covers a plethora of tools.
• Deciding the factors which help in choosing one way of instrumentation over the other.
• A way forward for organizations to evolve their instrumentation journey.

Reviewer info

This talk is primarily for practitioners who have to plan and implement the instrumentation journey for their organizations, the organizations that do not have the Observability system yet.

Audience Takeaways

Understand the landscape of instrumenting cloud-native applications to make informed decisions on how to instrument such software systems.

Speaker Profile

Prathamesh Sonpatki works at Last9.io building Reliability platforms for SREs. Contributor to open source and communities. More details at https://prathamesh.tech