End-to-end monitoring and automation for efficiency and performance gain
Use Cases and Deployment Scope
we monitor our energy management systems, assuring the realtime integrity of Big Data we generate from IoT devices and SCADA networks. It allows us to catch underperformance as soon as possible, reduce downtime, and ensure system availability. We use its APM and infrastructure monitoring for to enable energy audits, load forecasting, and keep our client facing analytics dashboards running smooth.
Pros
- New Relic APM allows us to follow up transactions across services and trace performance bottlenecks in real-time, crucial when monitoring the processing of energy loads or predictive maintenance algorithms.
- It gives us deep visibility into our cloud servers, containers and IOT gateways, so we can catch CPU spikes or memory leaks which can impact the data we ingest from the field devices.
- We develop custom dashboards for monitoring trends of power consumption, abnormality in sensors and API health. In conjunction with alerting, it makes sure we are fixing issues before customers even see them.
Cons
- And while powerful, building tailored dashboards with organ-specific metrics (such as energy load variance across regions) can be difficult to navigate. The UI isn't as drag-and-drop easy, and query-based widgets typically involve some trial and error for non-devs.
- Alerts may be hypersensitive or over general. I We often get a spam of non-critical alerts while doing load testing, all overhauling to me alone and making it difficult to identify actual issues especially in energy systems where spikes are very common.
- With our expanding fleet of Iot devices, the per-host pricing model is becoming expensive, quickly. More detailed billing based on microservices, or that works at sensor level, would make it more adaptable for energy platforms.
Likelihood to Recommend
It is perfect for observing our energy platforms during high-load situations, such as grid demand spikes, or our real-time ingest of sensor data, allowing us to respond in real time to anomalies. And it’s a good way to monitor API performance in client dashboards, too. But it’s not as conducive for low-budget/low-code work or deep customization that doesn’t get ample development support given its complexity and expense.
