Imagine standing in a massive library containing billions of books, documents, and images. Traditional search methods would have you searching for exact titles or authors. But what if you wanted to find materials with similar themes or content to a specific book? That's where vector search transforms the game—and it's now available in GreptimeDB.

The Limitations of Traditional Search

Conventional database search relies on exact matching or simple pattern matching:

• Keyword searches miss semantically similar content
• LIKE queries perform poorly at scale
• Fuzzy matching lacks true understanding of content

This approach breaks down entirely when searching for similar logs, related metrics patterns, or contextually relevant traces in your observability data.

Vector Search: The Semantic Revolution

Vector search transforms text, images, and other data into numerical vectors—essentially "digital fingerprints" that capture their meaning and characteristics. By measuring the similarity between these vectors, we can identify related content even when the exact wording differs.

How It Works in GreptimeDB

With v0.10, GreptimeDB integrated the powerful VSAG vector search library, bringing intelligent similarity search to your time-series data:

1. Embedding generation: Text or data is transformed into vector representations
2. Vector storage: These embeddings are stored in a dedicated VECTOR column type
3. Similarity search: Queries find data with embeddings most similar to the search target

The result? You can find logs with similar error patterns, identify metrics with comparable anomalies, or discover traces exhibiting similar behavior—even when the exact text differs.

Real-World Examples

Consider this practical example using the AG News dataset:

search_query = 'China Sports' 
search_embedding = model.encode(search_query)
query_statement = '''
    SELECT title, description, genre, vec_dot_product(embedding, :embedding) AS score
    FROM news_articles
    ORDER BY score DESC
    LIMIT 10
'''

Instead of returning only articles containing the exact words "China" and "Sports," this query intelligently retrieves articles about Chinese sports events, athletes, and related topics—delivering truly relevant results rather than rigid pattern matches.

Transformative Appl

Vector search opens new possibilities for observability platforms:

• Intelligent log correlation: Group semantically similar error messages across services
• Anomaly clustering: Find metrics exhibiting similar unusual patterns
• Root cause identification: Locate historical incidents with similar characteristics
• Natural language querying: "Show me services with connection timeout issues"

These capabilities represent a fundamental shift from reactive to proactive observability, where similar issues can be identified and addressed before they impact users.

Getting Started with Vector Search

Ready to explore the power of vector search in your observability stack? GreptimeDB makes it remarkably simple:

CREATE TABLE news_articles (
    title STRING,
    description STRING,
    embedding VECTOR(768),
    ts timestamp,
    PRIMARY KEY(title),
    TIME INDEX(ts) 
);

Combine this with any embedding model from libraries like sentence-transformers, and you're ready to start building intelligent search capabilities into your applications.

As systems grow more complex, the ability to find genuinely similar data will become essential. Vector search is no longer just a nice-to-have—it's the future of intelligent data querying in the observability space.

---

About Greptime

GreptimeDB is an open-source, cloud-native database purpose-built for real-time observability. Built in Rust and optimized for cloud-native environments, it provides unified storage and processing for metrics, logs, and traces—delivering sub-second insights from edge to cloud —at any scale.

• GreptimeDB OSS – The open-sourced database for small to medium-scale observability and IoT use cases, ideal for personal projects or dev/test environments.

• GreptimeDB Enterprise – A robust observability database with enhanced security, high availability, and enterprise-grade support.

• GreptimeCloud – A fully managed, serverless DBaaS with elastic scaling and zero operational overhead. Built for teams that need speed, flexibility, and ease of use out of the box.

🚀 We’re open to contributors—get started with issues labeled good first issue and connect with our community.

⭐ GitHub | 🌐 Website | 📚 Docs

💬 Slack | 🐦 Twitter | 💼 LinkedIn