Assets

Assets are different components identified by Digma in the tracing data. The represent the sum of unique elements in the application.

Asset categories include HTTP Endpoints, code locations, consumers, database queries, and more. In most cases, an asset correlates to a Span that has been categorized and processed in a specific way.

Another reason for categorizing spans into assets is to be able to better compare them. There is no sense in comparing an endpoint to a database query, for example. In each category, it is possible to sort the assets based on duration, performance impact, errors, and other criteria.

Asset Naming and Uniqueness

Assets are assigned unique names based on their role and attributes that are often different from the default Span name. This is a prerequisite to being able to group and aggregate data about them. This is why there isn't really a one to one mapping between spans and assets. another way to think about it is that a span is instanced in each trace while a Digma asset is extracted for multiple traces

Consider the following OTEL naming strategies vs. the name Digma assigned after some pre-processing.

Here are some examples:

Span Type	OTEL Naming	OTEL Example	Digma Naming	Digma Example
Database	SELECT <DBNAME> for example: SELECT 37b32f66-7a82-45f3-9d47-dc1d89b49ec7	SELECT 37b32f66-7a82-45f3-9d47-dc1d89b49ec7	Analyzes query syntax and generates a hash based query skeleton (without the parameters) .	select v1_0.pet_id,v1_0.id,v1_0.visit_date,v1_0.description from visits v1_0 where v1_0.pet_id=? order by v1_0.visit_date
HTTP Client	HTTP <VERB>	HTTP GET	Either HTTP <VERB> <Target Microservice Route> or HTTP <VERB> <Targer Domain>	HTTP GET /users_service/{userId}

Reviewing assets for your application or for specific APIs

The Digma Assets tab provides an overview of all recorded assets which were discovered in the traces. You can quickly search or filter to specific services or endpoints, and sort by duration, critical insights, and even performance impact.