Built in Seattle: RisingWave powers real-time at scale

While most databases still think in terms of batch processing and cloud-first architectures, Seattle-born RisingWave has quietly become the streaming database powering edge-first data layers for companies that can't afford the latency of traditional cloud-centric approaches. What started as a research project at the University of Washington has evolved into the go-to solution for real-time analytics that need to happen closer to users, not in distant data centers.

The hook — streaming SQL at the edge

RisingWave isn't just another database. It's a streaming SQL database that processes data as it arrives, making it perfect for edge-first architectures where computation needs to happen close to where data is generated. While traditional databases require you to move data to the cloud, process it in batches, and send results back, RisingWave lets you run continuous queries that update results in real-time — whether that's at AWS edge locations, on-premise servers, or even IoT devices.

This matters because Seattle's tech ecosystem has always been obsessed with scale and performance. From Amazon's massive cloud infrastructure to gaming companies like Valve and Bungie that need millisecond response times, local engineers understand that sometimes the cloud isn't fast enough.

Origin — from UW research to production reality

RisingWave emerged from the University of Washington's database systems group in 2021, led by researchers who had worked on stream processing at major tech companies. The team, including several engineers with backgrounds at Amazon and Microsoft, recognized a fundamental problem: as applications moved to edge computing, traditional databases were becoming the bottleneck.

"We kept seeing the same pattern," explains one of the original contributors. "Companies would build amazing edge applications, but then their data layer would force everything back through centralized cloud databases. The latency killed the user experience."

The project gained traction when local gaming companies started experimenting with it for real-time leaderboards and matchmaking. Unlike batch-oriented solutions that might update rankings every few minutes, RisingWave could maintain continuously updated leaderboards that reflected player performance in real-time.

Seattle's unique position as a hub for both cloud infrastructure (thanks to AWS and Azure's presence) and latency-sensitive applications (gaming, IoT, autonomous systems) created the perfect testing ground for edge-first database architectures.

Adoption — who's actually using it

Today, RisingWave runs in production at companies ranging from autonomous vehicle startups to financial trading firms. The common thread is applications where stale data isn't just inconvenient — it's catastrophic.

Local biotech companies use it for real-time analysis of lab equipment data, where traditional ETL pipelines would introduce dangerous delays in detecting equipment failures or contamination. Gaming companies leverage it for anti-cheat systems that need to detect suspicious patterns in player behavior within seconds, not minutes.

The project has also found adoption among Seattle's growing number of IoT companies. Traditional cloud-centric architectures require sending sensor data to centralized databases, processing it, and sending commands back — a round trip that can take hundreds of milliseconds. With RisingWave running on edge servers, that processing happens locally with sub-10ms response times.

Design choices that paid off

The RisingWave team made several architectural decisions that distinguished it from both traditional databases and other streaming systems:

SQL compatibility from day one

While many streaming systems invented their own query languages, RisingWave chose full PostgreSQL compatibility. This meant Seattle's database engineers could start using it immediately without learning new syntax or rewriting existing queries.

Elastic scaling without downtime

Unlike traditional databases that require careful capacity planning, RisingWave can scale compute resources up or down based on data volume. During Black Friday traffic spikes or gaming tournament peaks, it automatically allocates more processing power.

State management for edge deployments

The biggest challenge in edge-first architectures is maintaining consistent state across distributed locations. RisingWave's approach to distributed state management allows it to run reliably on everything from AWS edge locations to on-premise servers with intermittent connectivity.

Separation of storage and compute

By decoupling storage from compute, RisingWave can run computation close to users while still maintaining data durability in cloud storage. This hybrid approach gives companies the performance benefits of edge computing without sacrificing data safety.

One design choice that didn't initially work was trying to support every possible streaming use case. The team learned to focus on SQL analytics workloads rather than trying to compete with message queues or event streaming platforms.

How Seattle devs can contribute or learn from it

RisingWave is fully open source, and the Seattle tech community has been instrumental in its development. The project holds monthly meetups at various co-working spaces around South Lake Union, where contributors demo new features and discuss real-world deployment challenges.

For developers interested in contributing, the project needs help in several areas:

• Connector development: Building integrations with more data sources and sinks
• Performance optimization: Especially for edge deployment scenarios with limited resources
• Documentation: Real-world deployment guides for different edge computing platforms
• Testing: Load testing and chaos engineering for distributed edge deployments

Even if you're not ready to contribute code, studying RisingWave's architecture offers valuable lessons for any Seattle engineer working on distributed systems. The project's approach to handling network partitions, managing distributed state, and optimizing for different hardware profiles applies to many of the challenges facing local companies.

The project also demonstrates how Seattle's unique tech ecosystem — with deep expertise in both cloud infrastructure and latency-sensitive applications — can produce solutions that wouldn't emerge elsewhere.

For those working on edge computing projects, RisingWave provides a concrete example of how to architect systems that work well both in well-connected data centers and resource-constrained edge environments. The team's documented experiences deploying across different cloud providers and on-premise environments offer practical insights for similar projects.

FAQ

Q: How does RisingWave compare to traditional databases for edge deployments?

A: Traditional databases require moving data to centralized locations for processing, then moving results back. RisingWave processes data continuously where it's generated, eliminating the round-trip latency. For applications that need sub-100ms response times, this architectural difference is crucial.

Q: Can individual developers contribute to RisingWave without deep database expertise?

A: Absolutely. The project needs help with documentation, connector development, and testing — areas where general software engineering skills are more important than database internals knowledge. The monthly Seattle tech meetups often include RisingWave contributors who can help new contributors find appropriate starter issues.

The shift toward edge-first database architectures represents more than just a technical evolution — it reflects Seattle's broader understanding that the future of computing isn't just about moving everything to the cloud, but about intelligently distributing computation to where it's needed most.

---

Find Your Community

Interested in streaming databases, edge computing, or other cutting-edge data architectures? Connect with Seattle's vibrant tech community through our developer groups and discover upcoming events on distributed systems, database technologies, and more. Browse tech jobs from companies working on similar challenges, or check out tech conferences where you can dive deeper into edge-first architectures.