The IT Nerd

StarTree today announced support for Apache Iceberg in StarTree Cloud, enabling it to serve as both the analytic and serving layer on top of Iceberg, delivering interactive insights to internal and external applications directly from the data lakehouse. With this launch, StarTree redefines what’s possible with Iceberg, transforming it from a passive storage format into a real-time backend capable of powering customer-facing applications and AI agents with high concurrency serving thousands of simultaneous users with consistent speed and reliability.

While Apache Iceberg and Parquet have become popular open table formats for managing data in the lakehouse, they are not query engines and most existing query engines built around them struggle to meet the performance SLAs required for external-facing, high-concurrency analytical applications. As a result, companies have historically avoided serving data directly from their lakehouse, instead relying on reverse ETL pipelines or transforming the data into proprietary formats for separate serving systems—adding latency, complexity, and cost. StarTree removes these constraints by offering real-time query acceleration directly on native Iceberg tables. By combining open table formats like Parquet and Iceberg with Pinot’s powerful indexing and high-performance serving capabilities, StarTree enables applications to deliver live, interactive insights directly from the lakehouse without data duplication, format conversion, or operational trade-offs.

A Real-Time Serving Layer for Iceberg

StarTree Cloud integrates directly with Iceberg using open standards (Parquet and Iceberg table formats) and enhances performance with powerful indexing, intelligent materialized views (StarTree Index), and localized caching. Unlike traditional tools like Presto, Trino, or ClickHouse that rely on lazy loading and scanning, StarTree is built for low-latency, high-concurrency access, making it ideal for powering interactive dashboards, real-time data products, and operational workloads with strict SLAs.

Key capabilities include:

●        Native support for Apache Iceberg and Parquet in StarTree Cloud

●        Real-time indexing and aggregations, including support for numerical, text, JSON, and geo indexes

●        Intelligent materialized views via the StarTree Index

●        Local caching and pruning for low-latency, high-concurrency queries

●        No data movement required—serve directly from Iceberg

●        Intelligent prefetching from Iceberg, minimizing irrelevant data scans

With StarTree Cloud, companies can now unlock the full potential of their lakehouse investments and deliver modern, intelligent user experiences without architectural sprawl.

Availability
Support for Apache Iceberg in StarTree Cloud is available today in private preview. For more information, visit www.startree.ai.

Supporting Resources

●      StarTree Adds Native Iceberg Support: Serve High-Concurrency Queries Directly from Your Lakehouse

 StarTree today announced two new powerful AI-native innovations to its real-time data platform for enterprise workloads: Model Context Protocol (MCP) support and vector embedding model hosting. These capabilities enable StarTree to power agent-facing applications, real-time Retrieval-Augmented Generation (RAG), and conversational querying at the speed, freshness, and scale enterprise AI systems demand.

AI is only as powerful as the information architecture behind it. Just as the cloud forced a fundamental redesign of enterprise data systems—AI is now triggering a similarly profound shift. As agentic systems emerge, traditional data architectures—designed for internal users who accept slow queries and stale data—can no longer keep up. Agentic AI demands sub-second query speeds, real-time context awareness, and the ability to support swarms of autonomous agents working in parallel. This marks a fundamental shift in the role of data platforms—from static storage to dynamic engines that can aid agents in completing tasks.

StarTree has long delivered on this promise, powering millions of low-latency queries per second on the freshest data available. But new capabilities were needed to extend this foundation and fully unlock the next generation of AI-native applications. New features launching include:

• Model Context Protocol (MCP) support: MCP is a standardized way for AI applications to connect with and interact with external data sources and tools. It allows Large Language Models (LLMs) to access real-time insights in StarTree in order to take actions beyond their built-in knowledge. Availability: June 2025
• Vector Auto Embedding: Simplifies and accelerates the vector embedding generation and ingestion for real-time RAG use cases based on Amazon Bedrock. Availability: Fall 2025

The StarTree platform now supports:

• Agent-Facing Applications – By supporting the emerging Model Context Protocol (MCP), StarTree allows AI agents to dynamically analyze live, structured enterprise data. With StarTree’s high-concurrency architecture, enterprises can support millions of autonomous agents making micro-decisions in real time—whether optimizing delivery routes, adjusting pricing, or preventing service disruptions.
• Conversational Querying – MCP simplifies and standardizes the integration between LLMs and databases, making natural language to SQL (NL2SQL) far easier and less brittle to deploy. Enterprises can now empower users to ask questions via voice or text and receive instant answers—like a ride-hailing driver asking, “How much money have I made today?” followed by, “What about this month?” and “Where and when am I making the most money?”—with each question building on the last. This kind of seamless, conversational flow requires not just language understanding, but a data platform that can deliver real-time responses with context.
• Real-Time RAG – StarTree’s new vector auto embedding enables pluggable vector embedding models to streamline the continuous flow of data from source to embedding creation to ingestion. This simplifies the deployment of Retrieval-Augmented Generation pipelines, making it easier to build and scale AI-driven use cases like financial market monitoring and system observability—without complex, stitched-together workflows.

StarTree Expands Deployment Flexibility with Bring Your Own Kubernetes (BYOK)

StarTree also announced the general availability of Bring Your Own Kubernetes (BYOK), a new deployment option that gives organizations full control over StarTree’s high-performance analytics infrastructure within their own Kubernetes environments, whether in the cloud, on-premises, or in hybrid architectures.

With BYOK, enterprises can maintain full governance and control over their infrastructure while still taking advantage of StarTree’s real-time performance and ease of use. This model is ideal for regulated industries such as financial services and healthcare, where strict data residency, compliance, and security policies often prohibit the use of traditional SaaS models. It also delivers a cost-effective solution for organizations with stable, predictable workloads, offering savings on compute and egress fees.

BYOK joins StarTree’s existing deployment options, which include fully managed SaaS and Bring Your Own Cloud (BYOC), giving customers the flexibility to choose the model that best fits their operational and regulatory requirements. Availability: now in private preview

Real-Time Analytics Summit 2025: Coming May 14

StarTree will showcase many of these new innovations during the Real-Time Analytics Summit 2025, a virtual event taking place on May 14. The event will feature speakers from Uber, Netflix, AWS, and more, exploring the future of AI-driven analytics, data infrastructure, and emerging use cases across industries. Attendees will gain valuable insights into how real-time analytics is driving digital transformation across industries, from finance and e-commerce to gaming, cybersecurity, and beyond.

StarTree today announced it has been named the 2025 Confluent Data Flow ISV Partner of the Year – APAC. The award recognizes StarTree’s exceptional commitment to driving customer value through Confluent’s data streaming platform, alongside other global Confluent partners.

The Confluent Partner Awards for APAC recognizes regional partners that go above and beyond to deliver transformative customer value with data streaming–whether that’s through real-time business solutions or implementing cutting-edge technologies. The 10 regional award categories reflect the many ways partners across system integrations, cloud service providers, and technology partners leverage Confluent’s complete data streaming platform to connect, stream, govern, and process data as it happens.

StarTree provided outstanding services and solutions as the Data Flow ISV Partner of the Year – APAC. This award recognizes a partner that leveraged Confluent to create and deliver a comprehensive and compelling solution that made a significant impact across an industry and/or region.

StarTree and Confluent are a natural fit, seamlessly combining the strengths of real-time streaming and real-time analytics into a unified data platform. Both Apache Kafka® and Apache Pinot®, the open-source technologies respectively behind Confluent and StarTree, originated at LinkedIn to address the challenges of traditional batch-based data systems—enabling businesses to move from delayed insights to instant intelligence. Today, this partnership continues to redefine what’s possible with real-time data. With Confluent providing a best-in-class data streaming platform and StarTree delivering sub-second analytics at scale, organizations can unlock the full value of their data as it flows.

In 2024, StarTree consumed more data than any other real-time database natively integrated with Confluent Cloud. StarTree was also recognized as Confluent’s 2023 Integration ISV Partner of the Year, highlighting our sustained commitment to each other and the immense value we jointly bring to the market.

StarTree continues to thrive as a trusted and strategic partner in the channel, driving growth and innovation with its real-time analytics solutions. By offering seamless integrations with leading platforms such as Confluent, Tableau, AWS, Google Cloud, and Microsoft Azure, StarTree empowers its channel partners to deliver scalable and reliable insights that simplify complex business challenges. With a strong focus on collaboration, StarTree provides its ecosystem of hyperscalers, technology providers, and system integrators with the tools, resources, and expertise necessary to succeed in the rapidly evolving data landscape. Through flexible purchasing options in top cloud marketplaces and a commitment to building long-term relationships, StarTree ensures that its partners have everything they need to meet the dynamic needs of modern enterprises, ultimately delivering transformative value to customers worldwide.

Learn More about the StarTree + Confluent Partnership

Here’s some 2025 Predictions from Kishore Gopalakrishna, Cofounder and CEO, StarTree. This is what they see coming for next year that you should pay attention to.

The Dawn of Real-Time RAG for Dynamic Insights in 2025 – We’ll see the emergence of real-time Retrieval-Augmented Generation (RAG) as organizations push beyond batch processing limitations. Today’s RAG implementations primarily rely on static large language models (LLMs) paired with batch vector databases, which augment responses with preprocessed, stale data. While effective for many applications, this approach falls short for dynamic use cases that require real-time information updates, such as logistics optimization, personalized video game assistants, or financial risk monitoring. Real-time RAG will bridge this gap by integrating LLMs with real-time data streams and event-driven architectures, enabling models to access and leverage the freshest data during generation. This shift will unlock powerful, timely insights in scenarios where up-to-the-second context is critical, making 2025 a pivotal year for real-time augmented intelligence.

From Streams to Insights: 2025 Marks the Real-Time Analytics Revolution – Real-time analytics will finally hit its stride as organizations complete the “last mile” of their data architecture. Over the past few years, businesses have focused heavily on building out event streaming systems like Apache Kafka, ensuring that data flows smoothly in real-time. However, many are now realizing that traditional analytic endpoints, such as data warehouses and batch-based solutions, are unable to fully harness the potential of these streams. These legacy systems simply can’t deliver the instant insights needed in today’s fast-paced environment. In 2025, organizations will prioritize real-time analytics platforms that can process, analyze, and act on data instantly, closing the loop and unlocking the true value of their streaming architectures. This shift will enable innovative use cases such as hyper-personalized customer experiences, real-time external-facing data products, and adaptive risk management systems—far beyond the capabilities of traditional solutions.

2025 Will be the Year Observability Stacks Break Apart – Observability stacks are likely to become more disaggregated as companies move away from monolithic, all-in-one solutions to specialized, best-of-breed tools. As data volumes and complexity grow, teams will demand more flexibility in how they monitor and manage their infrastructure. This shift will result in observability stacks breaking into distinct layers—such as metrics, logs, traces, and events—each optimized with dedicated solutions. Disaggregation will enable more tailored observability strategies, greater scalability, and cost efficiency, as businesses can choose the most effective tools for specific parts of their systems rather than relying on a single, unified platform.