Request Metrics vs. Splunk Application Performance Monitoring (APM)

Request Metrics is great for small web development teams that want to make fast web applications. It was quick for us to get started and we could use the reports right away without having to read through documentation. If you just need to get some performance metrics, Request Metrics is perfect. It would not be so good for a larger team or a complex backend application as it is pretty limited in what it can measure.

Good for below cases 1. There is a front end and need to correlate data with front end data 2. multiple microservices and need to check the health of each system 3. correlate data from various sources 4. Application performance is a key to be captured 5. application performance is a key metric.

Incentivized

• Performance charts are really simple and easy for us to understand what is happening
• It was easy to get started and doesn't need administration or training
• It's affordable for our team
• It's fast

• Detect and Troubleshoot, which allows team to eliminate errors.
• Dashboard creates a one stop location for environment health
• Provides visibility, access to logs and user data real time.

Incentivized

• I wish that it could also do uptime checking of the site
• There is not enough detail to drill into individual users

• I wish Splunk Application Performance Monitoring could integrate with packet capture and analysis tools and provide the integrated analysis results on each tier of the application

Incentivized

Splunk is a great tool for log mining. It is less time-consuming and easy to use. It provides high security, it has customisable dashboards. Furthermore, it acts as a search head, and it gives real-time status. It has the ability to collect data from any source and multiple sources, and it can correlate them.

Incentivized

We also consider Speedcurve and Calibre for monitoring. Speedcurve was really nice, but the reports were slow. We had to wait a lot to get information. It was also out of our budget and would require extra justification to use it. Calibre was also nice, but the numbers were based on synthetic rather than real user performance. We felt that this wasn't as helpful for us.

It is good that the correlation analysis and the cause of the failure can be analyzed more extensively through the collection of many data. It is good because it provides a user interface (usability) well and can be used conveniently.
It is good because it can utilize anomaly detection through various AI-based ML models.

Incentivized

• The launch of our project dashboard was much smoother and more successful because we could fix problems earlier
• It fit without our existing operations budget

• It gave us more quicker alerts in a organize way to distribute to specific audience.
• Helped us to improve our uptime by getting pro-active alerts via emails
• More collaboration in a DevOps culture

Incentivized