Google recently introduced the new Google Cloud Inference API. The API allows developers to run large scale correlations over type time-series datasets? In operation, these correlations span a number of real world use case scenarios. Retailers can analyze foot traffic to sales conversion numbers using sensor data. Online retailers can track clicks, queries, and other UI interactions and correlate to responsive behavior. Google hopes to provide a single, simple method to run such correlations without the need to write custom scripts.
The API is currently in its alpha phase. Google is accepting applications to participate in the alpha. SnapChat has already integrated the API and sees its potential:
"Cloud Inference API promises to replace several custom dataflows with a single system that offers very low latency and real time updates," Peter Ciccolo, Snap Software Engineer, told Google. "Moreover, the same data can be used for multiple purposes, making it easy to explore new features with minimal custom code."
The API uncovers anomalies and detects trends within various datasets. The API is built to scale and can run tens of billions of events (thousands of queries per second). Because of its low latency, it can serve as a backend recommendation service for interactive experiences. The API is serverless and uses Google Cloud Storage to store the applicable time series data.