The emergence of AI and IoT is creating new opportunities for network ops infrastructure management, enabling prevention of outages, efficient use of resources and energy reduction. However, businesses today struggle to fully unlock these opportunities and maximize the value of data. Slow performance, costly manual management and limited data sources leading to limited insights are the current norm.
- In-memory speed
- Agile development
- Aggregation of a variety of data sources
- Continuous access to fresh data
- Hybrid and multi-cloud deployment
- Fine-grained security
- Volume and velocity: Performing real-time readings at scale, aggregated from multiple data sources concurrently
- Analysis: Limited machine learning capabilities and enrichment
- Complexity: The development and deployment of intelligent applications is complex, long and expensive
- Costs: Limited aggregation and filtering result in high storage costs
Using the Iguazio Continuous Data Platform, customers monitor and predict the health of provider edge and customer edge routers in the data center. The platform ingests data in high velocity from multiple streams, structured and semi-structured, at a rate of more than 50K events per second. Data from sources such as IBM Netcool, CA Spectrum and ServiceNow is aggregated in real-time and correlated with historical data. Conditional updates are used to cleanse the data before processing and atomic counters count the number of errors per router in the ingestion. By running Spark, abnormal events are detected with window-based computation and log messages are parsed automatically for context and severity of the failed component.
The system processes tens of thousands of messages with latency of seconds and serves the insights via interactive dashboards and alerts. The platform can be deployed at the edge, in the data center, in order to analyze the data closer to its sources and boost response time, or/and in the public cloud for elastic compute resources. Customers gain end-to-end visibility of network health with real-time detection, predictions and stats on previous failures. This is achieved with Iguazio’s simplified data pipeline which accelerates development and deployment, reduces the high storage costs of massive amounts of data and provides fine-grained security.