Fabric

Apache Flink is a distributed stream processing framework designed for real-time analytics and event-driven applications. It’s well known for its ability to handle large volumes of continuous data with stateful computation, meaning it can maintain context across many events - something that simpler streaming systems struggle with.

A key strength of Flink is its support for Complex Event Processing (CEP).

CEP allows you to define and detect patterns of events rather than just single occurrences. For example, you could detect a sequence such as three failed login attempts followed by a successful one, or identify correlated sensor readings that indicate a developing fault.

2026-10-12

Kafka Streams is a client-side stream processing library that operates directly on Kafka topics. Unlike a separate processing engine, it runs within your own application or service, allowing data transformation to occur close to where the events are produced.

This makes it ideal for preprocessing data before it reaches Fabric. You can use Kafka Streams to perform tasks such as data enrichment, filtering, or lightweight aggregation. For instance, before sending events to Fabric, you might:

2026-10-10

An Eventhouse-originally designed for realtime event data-can also serve as a vector database in Fabric RTI, supporting semantic/embedding-based retrieval workflows.

The key enabler is the dynamic column type, which allows you to store unstructured data such as arrays or JSON blobs. This lets your embedding vectors live alongside your ordinary event metadata.

Microsoft recommends using Vector16 encoding (i.e. 16-bit floating representation) rather than full 64-bit. This dramatically reduces storage footprint and boosts the performance of vector operations like dot product or cosine similarity.

2026-10-08

Retrieval-Augmented Generation, or RAG, is a technique where a large language model is given external context before it generates an answer. In Fabric, Eventhouse can play a central role in supporting real-time RAG solutions.

Because Eventhouse stores both structured and unstructured streaming data, it can serve as a continuously updated source of facts and events. Using built-in vectorization capabilities, recent data can be converted into embeddings and stored for semantic search.

2026-10-06

The Model Context Protocol, or MCP, provides a standard way for AI agents to interact with real-world systems and data sources.

RTI can expose real-time data and events through MCP-compatible endpoints.

When integrated with Fabric Real-Time Intelligence, MCP-based agents can consume real-time signals directly from Eventstreams, Eventhouses, or Activator outputs.

For example, an agent could subscribe to anomaly detection events and automatically create an incident ticket, adjust system parameters, or notify a human operator.

2026-10-04

As with any preview feature, there are some practical considerations.

The first is that this is still evolving. Not every scenario or data type is supported yet. Expect some limits on dataset size, refresh frequency, and concurrent analyses.

You must enable the Python plugin in the Eventhouse. This is what provides the model runtime, and enabling it isn’t instant. If you attempt analysis before it’s ready, the system will return an error.

2026-10-02

Let’s look at where this feature fits in practice.

The first major category is IoT telemetry - monitoring physical equipment. Think of a manufacturing plant where each machine sends readings for vibration, current draw, and temperature. Sudden spikes or drifts in these metrics may indicate wear, imbalance, or failure. The anomaly detector can continuously flag these irregularities before they cause downtime.

Another use case is application or system performance. If you’re monitoring throughput, error rates, or latency across multiple services, anomalies may indicate deployment issues, scaling problems, or outages. Because detection runs continuously and in near real time, operations teams can be notified within seconds.

2026-09-30

The AI Anomaly Detector in Microsoft Fabric Real-Time Intelligence uses a combination of time-series analysis and machine learning models. You don’t need to choose the model manually-Fabric evaluates the data and automatically selects the most appropriate approach.

Behind the scenes, it can use classical models like seasonal decomposition or ARIMA for steady, predictable series, or neural network–based models for more complex and irregular data. It also supports hybrid methods that adapt to both short-term trends and cope with long-term seasonality.

2026-09-28

The workflow is straightforward once you know the moving parts. You begin in your Eventhouse - that’s where data from your event streams is stored.

First, ensure that the Python plugin is enabled. This plugin provides the AI runtime used to train and run anomaly models. It can take up to an hour to enable for the first time, so plan that in advance.

2026-09-26

This new capability in Microsoft Fabric Real-Time Intelligence brings true AI-driven anomaly detection into the streaming environment. Traditionally, anomaly detection required a fair amount of data-science work - building models, testing parameters, and maintaining code. Fabric now handles much of that automatically.

The service examines incoming time-series or streaming data to look for unusual patterns - things that differ significantly from what’s been observed before. It supports both univariate detection, where you track a single metric such as temperature, latency, or transaction volume, and multivariate detection, which handles correlated signals - for example, temperature, pressure, and vibration from the same machine.

2026-09-24