Elevate Apache Pinot with StarTree Cloud
Support from the creators of Apache Pinot
Our team work hands-on with yours – from onboarding through production, helping with everything from data modeling and pipeline integration to query optimization and system tuning. Whether you need help scaling a critical dashboard, configuring ingestion from Kafka, or navigating schema design, you’ll have direct access to the same experts who helped companies like Stripe, Cisco, and DoorDash build their Pinot-based systems.
Fully managed, with flexible deployment options
SaaS
Get started quickly with the fully managed, StarTree Cloud service. Ideal for organizations that want zero ops overhead and rapid time-to-value.
Setup is simple, leaving you to focus on your data, your schema, and queries. And your SaaS instance is isolated from noisy neighbors at the infrastructure level.
BYOC
For customers who prefer dedicated infrastructure and deeper integration with enterprise systems, we offer a Bring Your Own Cloud (BYOC) model with VPC peering, custom SLAs, and enhanced observability.
In this case, the entire software stack is deployed in your cloud account and data never leaves your governance boundary
BYOK
For highly regulated environments, StarTree can be deployed behind your firewall—offering full control while still leveraging the power and capabilities of StarTree Cloud.
With this model you can deploy the entire stack in an existing Kubernetes cluster. This can be done in a given cloud provider (eg: Amazon EKS) or in a self-managed Kubernetes cluster on baremetal.
StarTree enhances Apache Pinot with features that make it more powerful while lowering costs.
Interactive Queries with Tiered Storage
StarTree’s “Precision Fetching” on tiered storage enables Pinot to do fast, interactive queries on data stored in cloud object stores (like S3). This capability is highly performant as it is able to fetch only the data needed for the query.
Precise fetch makes it viable to move historical data to more affordable cloud storage – all while maintaining interactive query response times.
Upserts that scale
Upserts in open-source Apache Pinot are limited by the memory overhead needed to handle the primary key map. This puts a limit on how far you can push upsert functionalities.
StarTree Cloud has addressed this limitation to allow updates to billions of primary keys per server—without compromising speed or concurrency. This makes real-time, mutable datasets possible at scale.
Data management made simpler and more flexible.
Automate operations on StarTree Cloud
Performance Manager
Schema Evolution
Data Backfill
Dynamic Autoscaling
StarTree enables you to dynamically scale and optimize resources with Minion Autoscaling to avoid paying for idle infrastructure while maintaining performance under load.
Role Based Access Control
Improved Query Console
Ready for AI and modern workloads
Query data in Iceberg & Parquet
Query directly from your data lakehouse. StarTree’s Iceberg and Parquet support opens up the ability to serve low latency and high concurrency analytics without moving data or stitching together brittle pipelines.
MCP Server for Apache Pinot
MCP provides a standardized interface for AI models to interact with your data. StarTree’s MCP Server enables AI applictions to get instantaneous responses to queries on rapidly changing data – critical for maintaining AI agent context and fluid interactions
Vector auto-embedding for building real-time RAG.
Apache Pinot added support for vectors in 2024. This enables AI models to reason over live data instead of stale snapshots. And with vector auto embedding, StarTree makes that power easier to harness. Generate, ingest, and index vector embeddings in a fully integrated workflow, bypassing the need for complex, stitched-together pipelines.
Grafana integration to support observability
StarTree offers seamless integration with Grafana for data visualization and monitoring. This integration allows users to leverage Grafana’s dashboards to interactively visualize data stored and analyzed by Pinot.
Your data remains private and secure
Robust Authentication & Authorization
Encryption of data at rest and in transit
Networking Security
Audit trails and data loss prevention
Real-time use-cases thrive on StarTree
Token-Level Observability for LLMs: How Together AI Does It
As large language models (LLMs) become core infrastructure, a new engineering challenge is taking shape: capturing how these models are used—and how they behave—as it happens. It’s no longer sufficient to count API calls or log failures after the fact. …
MiQ Reinvents Programmatic Campaign Building with a Unified, AI-Driven Audiences Platform
MiQ’s Audiences engine is powered by a modern data stack designed for speed, scale, and flexibility. MiQ replaced Amazon Athena with StarTree, a real-time analytics engine built on Apache Pinot, to support the platform’s advanced search and indexing capabilities.
Analyzing Financial Transactions Using StarTree Cloud
Razorpay – India’s fastest-growing payment processing company – chose StarTree, powered by Apache Pinot, for real-time analytics in its user-facing applications, including its Success Rate and internal monitoring dashboards.
How do Pinot and StarTree Compare?
| Group | Feature | Open Source Apache Pinot | StarTree Cloud Adds… |
|---|---|---|---|
| Managed Service | Operations & Management | Users manage cluster tuning, scaling, upgrades, monitoring | Fully managed by Pinot creators; automated lifecycle mgmt |
| Support | Community-based support | Proactive monitoring and management by the largest team of Pinot committers/PMCs, with enterprise-grade SLAs. More | |
| System Observability | Limited | Out of the box support for system metrics (Grafana), logs (Loki) and query debuggability (Loki + StarTree Query Console) | |
| Value-Add Features | Minion Auto Scaling | Not available | Automated background tasks (compaction, rebalancing, indexing) without query impact. More |
| Precise Fetching | Not available | Retrieves only needed data from S3, delivering sub-second queries while cutting compute/storage costs. More | |
| AZ-Aware Kafka Ingestion | Custom Setup | Out of the box setup to reduce cross-AZ traffic costs during ingestion | |
| AZ aware HA setup | Custom setup | Out of the box support for multi-zone deployments for HA deployments. More | |
| Scalable Upserts | Limited to in-memory | Enterprise-grade off-heap upserts with high-concurrency and fast updates. More | |
| Advanced Indexes | Core indexes only | Sparse Index (efficient selective queries); Composite JSON Index (nested queries) | |
| ML-Driven Query Optimization | Manual tuning required | Automated index selection and query tuning. More | |
| Automated Data Backfill | Manual | Detects and fills data gaps without pipeline disruption. More | |
| Security | Basic controls | Role-based access control, encryption, SSO (OIDC), SOC2/ISO/HIPAA compliance, audit logs and AWS secrets manager support. More | |
| Data onboarding Wizard | Limited | Intuitive UI based wizard for onboarding new datasets with automated schema inference, data preview and data modelling support | |
| Integrations | Apache Iceberg | Not built-in | Native Iceberg connector for lakehouse queries without pipelines/duplication. More |
| Delta Lake | Not built-in | Native Delta Lake connector. More | |
| Snowflake | Not built-in | Native Snowflake connector. More | |
| Grafana | Not built-in | Native plugin + PromQL-style queries for observability. More | |
| Model Context Protocol (MCP) | Not available | Secure, real-time LLM/AI agent access to Pinot data. More | |
| Vector Embedding Hosting | Not available | Hosts embeddings for semantic search & RAG use cases.More | |
| AWS Graviton Support | Not optimized | Cost-efficient compute with ARM-based Graviton processors.More |
Request a Trial
The best way to discover real-time analytics with Apache Pinot is to try it yourself – and there’s no easier way than StarTree Cloud.
