top of page
Refresh a DB Performance Report Template

AWR Report

Automatic Workload Repository (AWR) report is a one- page performance report for Oracle databases. It collects and stores performance statistics for problem detection and self-tuning. AWR us used by a broad audience both internally and externally. However, it hasn’t gotten any design love since it was first released in early 2000. Our VP asked for a refresh.




Role / 

UX/UI Designer


Collaboration / 



Year / 

Jul, 2023- Sept, 2023


Modern AWR, What Does it Mean?

Me and the other designer Val in the team were tasked to refresh this report. However, our VP left it open for us to define what the next-gen AWR should look like .


Our first step with this project is to conduct foundational research and make our own definition on a modernized AWR report.

Information Gathering

Getting the Basics

Before user research, we collected some information from our VP and supporting team. 

What are AWRs?

We were shared with a bundle of AWR reports. Turned out there are more than one types of AWR reports.


Each team at Oracle focuses on using different types of AWR reports. These reports' formats are similar, but data generated varies from one report type to another. 

MacBook Pro 14_ - 1.png

Who are the report users?

Internal: Domain Experts

Internally, this report is used by various teams, including bug diagnosis for customer support, performance analysis and feature testings.

External: DBAs

Customer can download this report from OCI console to do self diagnosis, or pass it down to Oracle support team to identify bugs. 

User Research

User Research

More Things to Know

This poses a design challenge, how do we modernize the report in a way that it can be applied to a variety of reports used by different audience?

To tackle this, Val and I worked together and identified a few things we want to investigate.

Know More About AWR Reports and Users

What are the teams using AWR report?

How many types of AWR reports are there?

Which team manages the data generation?

Understand User Journey

How is AWR used?

Is there any common user journey, or common section that’s consistently shown in all types of report.

Discover Design Opportunities

What’s working and what’s not working with the current report?


We talked to 5 internal AWR users, each from different team. A round of affinity diagraming leads to the following findings. 

Frame 91799.png
Frame 91799.png

More About AWR Reports and Users

User Journey

Internal domain experts typically have an idea of what they're looking for. They look at key metrics under report summary, then depending on their focus, they will go to sections like wait events, Exadata, or SQL to branch down.

MacBook Pro 14_ - 2.png

Design Opportunities

😃 What’s working?​

Quality of data is good so far

  • Quality of data = data users needed to identify a root cause of a problem is available in the report.

  • There is more data in the report than needed for any given problem, but it’s preferred that way as each problem being solved requires different data.

  • AWR data is iterative and keeps expanding each release, internal users also have channels to give feedback and add data in the report.

Frame 356.png

😞 What’s not working?​

Presentation is secondary, but there's a lot to improve

  • Users would like better navigation and readability.

  • Wish there's a way to trim down noise once user got an area of focus. User only look at a small section or certain metrics once the issue is identified. With the current report, they rely on other tools to copy/paste related data and put them together, or they go back and forth to cross reference.



Information Architecture (IA)

Amount of data has grown over time, but in a way that looks like things being simply piled up on top of each other.Its information architecture reflected this.


The first thing we did is to understand the existing IA so we can come up with a solution to re-organize information for better navigation.

Frame 91804.png

Intro section: gives context of the environment

Report summary: key metrics

Report index: index is in between the context which is a big usability issue

Content: tremendous number of tables and content

L1 navigation

L2 navigation


Frame 91800.png

Index (anchor links)







Frame 20311.png

From IA to Design

The existing report has 5 levels of navigation.

L0- (eg.) Main Report 

L1- (eg.) Report Summary

L2- (eg.) Load Profile

L3- (eg.) Exadata outlier summary

L4- (eg.) Cell level

In design, the L4 level is not shown in the index as this is too granular.

Content in the Report

Today, data in AWR are in grid and raw text format.

In this wireframe, we defined the boundary and put related sections together so the IA aligns with the index.
Screenshot 2023-09-16 at 4.57.17 PM.png

Define Component Behavior

To improve readability and make sure it's scalable, guideline and behavior of the component and defined.

The grid component is from Oracle's redwood design library. A few customization is made so that it's compact, readable, and it can be scaled throughout the report. 
Screenshot 2023-09-16 at 5.51.33 PM.png
Grid component.png

Additional Features- Comments

Users today already use email or txt file to document their findings and share with their co-workers or customers. 

We added this built in comment feature, which allows for a quick navigation between related info, trim down noise, and great increate productivity.
Screenshot 2023-09-16 at 6.04.50 PM.png

Final Design

Descriptions on.png


bottom of page