Observability in Multi-Cloud Environments

Study Notes

Jeff, a Cisco Lab Dynamics manager, and Sunny Lu, a Cisco product management director, discuss the challenges of observability in today's multi-cloud environment.
Multi-cloud is a prevailing trend, with companies like AWS, Google, and Azure growing rapidly, making it complex for customers to observe all endpoints.
In a multi-cloud world, applications are becoming distributed and highly complex, consisting of on-premises, cloud, and edge components.
Traditional applications still exist, but modern applications are evolving, with things becoming serverless and microservices.
Customers have been living in silos with various tools designed for specific domains, such as logging, metrics, and tracing.
Opening up data centers and applications to multi-cloud introduces dependencies, making it crucial to have a full view of the application from a business perspective.
Sunny mentions the concept of "Shadow IT," explaining how multi-cloud environments have evolved organically rather than strategically.- The text discusses the need for observability and data governance in cloud environments, with a focus on the open Telemetry framework.
Shadow ID was once seen as a problem to be stopped, but is now seen as a necessary compulsion due to the prevalence of cloud practices and the movement of data.
Cloud practices teams in organizations now determine how security can be implemented for cloud practices, as data is moving around frequently and visibility is critical.
Multiple tools are often used by customers to address their needs, but they complain about having to manage various data formats, proprietary data, and multiple agents.
Open Telemetry is a solution to this problem, allowing observability data such as metrics, events, and traces to be collected in a single collector and sent to any open Telemetry backend for analysis.
Open Telemetry enables transformations like visualization, percentiles for metrics, and breaking down spans into multiple traces.
The importance of collecting data once and using it multiple times for different set of use cases or personas to save resources is highlighted.
The FSO platform is described as a full stack observability data platform born with the need for full stack observability.
The platform is geared towards making business decisions based on various industry needs and allows for a common context, single agent, and cost savings.
App Dynamics, a Cisco acquisition, was the first application within the Cisco portfolio of FSO and was created to go beyond application performance management to look at infrastructure and network issues.
The platform was built with an architecture that allows it to be used Cisco-wide and industry-wide as a standard for observability.
Open Telemetry is the underpinning architecture of the FSO platform, simplifying data collection, standardization, and making it easy for all personas to get the same data in different tools.
The FSO platform is not about creating products that can only be used for Cisco-specific use cases but about creating a platform for an ecosystem of developers to build applications for observability.
The platform allows the definition of relationships between any two things that are being monitored, including devices and sensors emitting data.
Open Telemetry enables the platform to take any data type and convert it into open Telemetry, allowing for the creation of context and relationship building between entities.
The FSO platform provides the flexibility for developers to extend what is provided and solve customer use cases rather than waiting for something from Cisco.

Description

Explore the challenges and solutions for observability in today's multi-cloud environment, including the impact of distributed applications, the concept of Shadow IT, the adoption of Open Telemetry framework, and the development of the FSO platform. Learn about the evolving nature of cloud practices and the importance of full stack observability.