At VeeamON 2024, I had the pleasure of introducing our latest innovation: the next generation Veeam Intelligence integrated within VeeamONE. This isn’t just any ordinary chatbot, it’s a shift in mindset for backup administrators, application owners, and C-level executives alike. By leveraging the power of real-time insights through our powerful APIs, this AI assistant provides a new way to do analytics, reporting and support.

This is true natural language analytics and observability powered by real data from the Veeam Data Platform

A Chatbot on Steroids

Our new Veeam Intelligence seamlessly integrated into the VeeamONE console. It’s designed to provide real-time insights and analytics by tapping into the VeeamONE API, making it a powerful tool for users who need quick, accurate information without diving deep into the backend.

The Technical Magic: RAG and VeeamONE APIs

So, how exactly does our Veeam Intelligence achieve this level of functionality? It’s all about the integration of Retrieval-Augmented Generation (RAG) and VeeamONE APIs.

Retrieval-Augmented Generation (RAG):

RAG is an approach that combines the strengths of information retrieval and generative models to provide accurate, contextually relevant responses. Here’s a deeper dive into how it works:

  • Information Retrieval: The AI assistant uses information retrieval techniques to pull relevant data from VeeamONE APIs in real-time. This ensures that the information is always up-to-date and reflects the current state of the environment.
  • Contextual Understanding: The retrieved data is then processed by a natural language understanding (NLU) component, which interprets the context and nuances of the user’s query. This step is crucial for generating meaningful responses that address the specific needs of the user.
  • Generative Response: Finally, a generative model creates a detailed, contextually appropriate response. This model is trained on vast amounts of data to understand how to construct responses that are not only informative but also engaging and easy to understand.

VeeamONE APIs:

The VeeamONE APIs are the backbone of this integration. They provide the necessary endpoints for accessing various data points within the VeeamONE environment, such as:

  • Alarms and Events: The API allows the AI assistant to retrieve alarm details, including unusual jobs and security vulnerabilities. This enables real-time monitoring and immediate action.
  • Threat Analysis: By querying the Threat Center API, the assistant can gather and summarize critical threat information. This helps in identifying potential security risks quickly.
  • Backup Job Details: The API facilitates the retrieval of backup job statuses, restore points, and server protection information, ensuring that all critical backup processes are tracked and managed effectively.
  • Server Management: The assistant can list server versions and provide upgrade recommendations based on the data pulled from the API, making it easier to maintain up-to-date systems.

Contextual Analysis and Follow-Up:

Once the data is retrieved, the AI assistant employs natural language processing (NLP) to analyze the context and generate appropriate follow-up actions. This includes:

  • Identifying Critical Alarms: The assistant assesses alarm data to highlight the most critical issues and suggests immediate actions.
  • Summarizing Threat Data: It condenses extensive threat analysis into a concise summary, making it easier for administrators to understand and act upon.
  • Providing Detailed Insights: For specific queries, like the protection status of SQL servers, the assistant drills down into the data to present detailed insights.
  • Suggesting Next Steps: Based on the retrieved data, the assistant offers actionable recommendations, such as server upgrades or security enhancements.

Demonstration Highlights

During the demo, I showcased several key functionalities of Veeam Intelligence:

  • Alarm Overviews: I started with an overview of alarms within the environment. The AI assistant pulled real-time data, highlighting unusual jobs and security vulnerabilities. It then provided a contextual response, identifying the most critical alarms and offering recommended actions.
  • Threat Center: The Threat Center feature, introduced in the latest product version, was next on the list. The AI assistant accessed the Veeam API to deliver real-time insights, summarizing vast amounts of data into concise, actionable information.
  • SQL Server Protection: I demonstrated how the AI assistant can help a backup admin verify the protection status of SQL servers. By querying the system, it quickly identified protected servers and provided detailed information about their restore points.

Wrap Up

This demonstration at VeeamON 2024 highlighted the transformative potential of the next generation AI assistant in VeeamONE. This innovative tool simplifies complex tasks, enhances decision making, and paves the way for more advanced AI integrations. By leveraging RAG and VeeamONE APIs, we provide real-time data access and contextual analysis that revolutionize how users interact with their data.

Stay tuned for more updates and advancements as we continue to push the boundaries of what’s possible with Generative AI in data management.

Full VeeamON Mainstage Demo here: https://youtu.be/cTTWshTIY_Y?si=4jlE4mMzUvokaFwj&t=3104