Real-time OLAP databases are optimized for speed and often rely on tightly coupled storage-compute architectures using disks or SSDs. Decoupled architectures, which use cloud object storage, introduce an unavoidable tradeoff: cost efficiency at the expense of performance. This makes them unsuitable for databases that need to provide low-latency, real-time analytics, especially the new wave of LLM-powered dashboards, retrieval-augmented generation (RAG), and vector-embedding searches that thrive only when fresh data is milliseconds away. Can we achieve both cost efficiency and performance?
In this talk, we’ll explore the engineering challenges of extending Apache Pinot—a real-time OLAP system—onto cloud object storage while still maintaining sub-second P99 latencies.
We’ll dive into how we built an abstraction in Apache Pinot to make it agnostic to the location of data. We’ll explain how we can query data directly from the cloud (without needing to download the entire dataset, as with lazy-loading) while achieving sub-second latencies. We’ll cover the data fetch and optimization strategies we implemented, such as pipelining fetch and compute, prefetching, selective block fetches, index pinning, and more. We'll also share our latest work about integration with open table formats like iceberg, and how we will continue to achieve fast analytics directly on parquet files by implementing all the same techniques that apply to tiered storage.
Real-time OLAP databases are optimized for speed and often rely on tightly coupled storage-compute architectures using disks or SSDs. Decoupled architectures, which use cloud object storage, introduce an unavoidable tradeoff: cost efficiency at the expense of performance. This makes them unsuitable for databases that need to provide low-latency, real-time analytics, especially the new wave of LLM-powered dashboards, retrieval-augmented generation (RAG), and vector-embedding searches that thrive only when fresh data is milliseconds away. Can we achieve both cost efficiency and performance?
In this talk, we’ll explore the engineering challenges of extending Apache Pinot—a real-time OLAP system—onto cloud object storage while still maintaining sub-second P99 latencies.
We’ll dive into how we built an abstraction in Apache Pinot to make it agnostic to the location of data. We’ll explain how we can query data directly from the cloud (without needing to download the entire dataset, as with lazy-loading) while achieving sub-second latencies. We’ll cover the data fetch and optimization strategies we implemented, such as pipelining fetch and compute, prefetching, selective block fetches, index pinning, and more. We'll also share our latest work about integration with open table formats like iceberg, and how we will continue to achieve fast analytics directly on parquet files by implementing all the same techniques that apply to tiered storage.
Videos:
Presentation Slides:
Real-time OLAP databases are optimized for speed and often rely on tightly coupled storage-compute architectures using disks or SSDs. Decoupled architectures, which use cloud object storage, introduce an unavoidable tradeoff: cost efficiency at the expense of performance. This makes them unsuitable for databases that need to provide low-latency, real-time analytics, especially the new wave of LLM-powered dashboards, retrieval-augmented generation (RAG), and vector-embedding searches that thrive only when fresh data is milliseconds away. Can we achieve both cost efficiency and performance?
In this talk, we’ll explore the engineering challenges of extending Apache Pinot—a real-time OLAP system—onto cloud object storage while still maintaining sub-second P99 latencies.
We’ll dive into how we built an abstraction in Apache Pinot to make it agnostic to the location of data. We’ll explain how we can query data directly from the cloud (without needing to download the entire dataset, as with lazy-loading) while achieving sub-second latencies. We’ll cover the data fetch and optimization strategies we implemented, such as pipelining fetch and compute, prefetching, selective block fetches, index pinning, and more. We'll also share our latest work about integration with open table formats like iceberg, and how we will continue to achieve fast analytics directly on parquet files by implementing all the same techniques that apply to tiered storage.
Complete the form below to access the full overview:
.png)
Videos
Unlock the full performance of your AI/ML infrastructure on Oracle Cloud Infrastructure (OCI).
Join Oracle's Master Principal Cloud Architect Xinghong He and Alluxio's VP of Technology Bin Fan for an in-depth technical session exploring how modern tiered caching, optimized storage integration, and smart deployment choices can deliver sub-millisecond latency and up to 5× faster data access on OCI — at scale.
You'll learn about:
- Architectural insights: How Alluxio’s tiered caching architecture works with OCI Object Storage and BM.DenseIO compute instances to eliminate data access bottlenecks.
- Benchmark-proven results: See real MLPerf Storage 2.0 and Warp benchmark outcomes demonstrating sub-millisecond latency and dramatic throughput gains.
- Deployment strategies: Compare deployment options — dedicated mode for peak performance vs. co-located mode for cost-efficient scale.
- Practical, actionable guidance: Implementation best practices you can apply directly to your AI/ML workloads on OCI.
Fireworks AI is a leading inference cloud provider for Generative AI, powering real-time inference and fine-tuning services for customers' applications that require minimal latency, high throughput, and high concurrency. Their GPU infrastructure spans 10+ clouds and 15+ regions, serving enterprises and developers deploying production AI workloads at scale.
With model sizes reaching 70GB+, Fireworks AI faced critical challenges: eliminating cold start delays, managing highly concurrent model downloads across GPU clusters, reducing tens of thousands in annual cloud egress costs, and automating manual pipeline management that consumed 4+ hours weekly. They chose Alluxio as their solution to scale with their hyper-growth without requiring dedicated infrastructure resources.
In this tech talk, Akram Bawayah, Software Engineer at Fireworks AI, and Bin Fan, VP of Technology at Alluxio, share how Fireworks AI uses Alluxio to power their multi-cloud inference infrastructure.
They discuss:
- How Fireworks AI uses Alluxio in its high-performance model distribution system to deliver fast, reliable inference across multiple clouds
- How implementing Alluxio distributed caching achieved 1TB/s+ model deployment throughput, reducing model loading from hours to minutes while significantly cutting cloud egress costs
- How to simplify infrastructure operations and seamlessly scale model distribution across multi-cloud GPU environments
