🚀 Apache Iggy: From Rust Experiment to Apache Top-Level Project

Date: August 24, 2026
Authors: Piotr Gankiewicz (Founder & PMC Member) and Kranti Parisa (PMC Chair)

We are thrilled to announce a major milestone: as of August 19th, 2026, Apache Iggy has officially graduated from the Apache Incubator. Following a unanimous positive vote, Iggy is now a full-fledged Apache Software Foundation Top-Level Project (TLP).

What began over three years ago as a modest exploration into the Rust programming language and the mechanics of message streaming has evolved into a global, independent project. Remarkably, the transition from the Incubator to TLP status took only about 18 months.

---

🛠️ The Origins: Driven by Curiosity

The story of Iggy started in March 2023. It wasn't born from a corporate requirement or a gap in the market, but rather from Piotr Gankiewicz’s personal desire to master Rust and understand the inner workings of messaging systems.

While the industry already had established giants like Apache Kafka, the drive to build something new was fueled by a classic engineering obsession: latency, efficiency, performance, and operational simplicity.

"Engineers still build new database engines all the time... driven by a simple 'let's just build what we desire' attitude."

The project was named Iggy, after the Italian Greyhound—a breed known for being small yet possessing unmatched speed.

The Technical Evolution

Initially, the goal was basic: create a simple append-only log server. However, the pursuit of performance is a "deep rabbit hole." The project evolved from a solo side project to be completed in three weeks into a sophisticated streaming platform.

Current Technical Architecture:

• Design: Thread-per-core architecture.
• I/O: Utilizes io_uring for high-efficiency disk and network operations.
• Consensus: Implements VSR (Viewstamped Replication Revisited).
• Performance: Achieves single-digit millisecond latencies, expressed as: $\text{Latency}_{P99+} < 10\text{ms}$

// A conceptual representation of the initial append-only goal
struct IggyLog {
    storage: Vec<u8>,
    offset: usize,
}

impl IggyLog {
    fn append(&mut self, data: &[u8]) {
        self.storage.extend_from_slice(data);
        self.offset += data.len();
    }
}

---

🌐 Building a Global Community

The project shifted from a personal experiment to a community effort the moment external developers began contributing via pull requests, bug reports, and technical discussions.

• Discord: Became the hub for real-time interaction, brainstorming, and Q&A.
• GitHub: Served as the transparent ledger for technical decisions and development.

The Path to the Apache Incubator

To ensure the project's long-term viability and neutrality, the team sought a governance model where no single person or company held absolute control. In early 2025, the project entered the Apache Incubator.

Key Transition Details:

1. Ownership: LaserData Inc. transferred the source code to the ASF via the Software Grant process.
2. Governance: Shifted to a community-led model based on public votes, consensus, and the PMC.
3. Leadership: Kranti Parisa guided the onboarding and graduation process and now serves as the Chair of the Iggy PMC. (Note: The Chair acts as a liaison to the ASF Board rather than a "boss," as the PMC governs collectively).

Special Acknowledgments:

• Yonik Seeley: Apache Champion and Mentor.
• Hulk Lin, Zili Chen, and Hao Ding: Mentors who provided critical guidance on releases and governance.

---

📈 Growth Metrics

During its 18-month incubation period, Iggy's growth was exponential.

While these numbers are impressive, graduation is primarily a testament to the project's health, diversity, and self-governance.

---

🗺️ The Road Ahead

The journey from the first line of code to a TLP took less than three and a half years. Now, the community is focusing on several critical technical milestones:

Project Roadmap

• Clustering: Release the next version featuring VSR-based clustering for high availability.
• Interoperability: Finalize the initial Kafka protocol gateway to integrate with existing ecosystems.
• Research: Explore the implementation of multi-leader replication.

The full list of PMC members, committers, and mentors can be found on the official Team page.