How CS Corp Uses the Use Case Explorer for Observability
Who is CS Corp?
CS Corp is a fictitious corporation whose sales are conducted online through their virtual reality web store called the Online Boutique. They are a new entrant in their market and will be delivering an innovative new user experience using cloud-native technologies.
Online user experience is everything in this environment, and CS Corp's competitors have no idea that they are working on this revolutionary new initiative. The web store consists of an Amazon EC2 infrastructue environment and cloud-native application services running in Kubernetes managed containers.
CS Corp's marketing team recently conducted a market study of beta customers and had some surprising feedback. Customers love the experience up until they go to checkout. They said that the site randomly becomes too slow to continue the purchase. Several even commented that because of the experience, they would probably search competitive sites for a similar product. Marketing approached their application team and site reliability engineer (SRE) and informed them of this feedback.
With only a short time before the huge market launch, they requested the application performance issue be fixed immediately. In addition, they requested that the SRE also benchmark site performance against their competitor. The marketing team also questioned why the experience is inconsistent. The SRE told the marketing team that he was periodically checking logs and performance information manually. If he saw anything concerning, he remediated the issue or contacted either an infrastructure team or the application developers.
This is not an ideal operating scenario for CS Corp. Let's see how the Use Case Explorer can be used to solve this problem.
How CS Corp Uses the Use Case Explorer for Observability
The team at CS Corp meet to discuss the problem. Looking at the Value Realization Cycle as a guide, they begin to plan out the work that will be required to solve the problem. Here are the steps they take.
1. Define the problem and goal.
The Online Boutique is performing slowly in an inconsistent manner which frustrates customers to the point that it may drive them to competitors to purchase their items. We also need to benchmark our competitor's experience from a performance perspective to ensure we are outperforming them, which is critical in this competitive market.
2. Identify and record use cases.
Using the Explorer Map, we determine that we require visibility into the Infrastructure, Application, and Digital Experience layers in the Observe stage. We also determine that in order to optimize root cause analysis (RCA) and automate notification of the SRE, Infrastructure team or Application Developer, we will want to leverage use cases in the Engage stage. Finally, in order to speed resolution when issues arise, we also want to look at the Incident Investigation use cases in the Act stage.
3. Deploy use case(s) and document the value achieved.
CS Corp records the use cases it wants to complete in the Use Case Registry.
The team also sets up a weekly team meeting focused on the Use Case Registry to checkpoint and track progress. In addition, they identified where they could use Splunk OnDemand credits to help deploy a couple of use cases and speed up their overall implementation time.
Here is the Registry CS Corp starts to work with. Over time, they add to the Registry with updates on the completion of use cases, as well as new use cases that have been identified through their monthly use case planning sessions.
| Workflow Stage | Category | Use Case | Product | Expected Value | Owner | Target Date | OnDemand Credits |
|---|---|---|---|---|---|---|---|
| Observe | Digital Experience | Monitor user response time | Synthetics | Understand the customer experience via SLI | Michelle Jackson | July 15 | |
| Observe | Application | Observe function call trace time | APM | Reduce MTTD | Michelle Jackson | July 20 | |
| Observe | Infrastructure | Monitor Kubernetes pods for pending state | Infrastructure Monitoring | Reduce MTTD | Jack Handley | July 15 | |
| Observe | Infrastructure | Monitor AWS EC2 availability | Infrastructure Monitoring | Reduce MTTD | Jack Handley | July 15 | 10 |
| Engage | Event Analytics | Identify causal component so that notification can be routed to correct support team | ITSI | Reduce mean time to isolate | Siraj Chaudry | July 25 | 10 |
| Engage | Notification | Send notification to responsible support team | OnCall | Reduce mean time to notify | Siraj Chaudry | Aug 9 | |
| Act | Incident Investigation | Train how to investigate application trace time issues | APM | Reduce mean time to investigate | Michelle Jackson | Aug 9 | |
| Act | Remediation | Train how to roll back releases | Multiple | Reduce mean time to remediation | Michelle Jackson | Aug 12 |
Now CS Corp has a roadmap of the use cases they need to accomplish their objectives, the products they need to use, and the expected value and owners. In addition, they've identified where they could use their Splunk OnDemand credits to help deploy a couple of use cases and speed our overall implementation time.
Next, they can search the Use Case Observability Library to see what other use cases could help them quickly accomplish their objectives.