ez-cdc
    Built with Rust • Single Binary

    Real-time CDC without the pain

    ez-cdc is a Rust-based Change Data Capture platform that replicates PostgreSQL to modern analytics platforms with sub-second latency — a single binary with no JVM, Kafka, or ZooKeeper.

    View on GitHubRead the Docs
    $ez-cdc quickstart starrocks

    ✓ source connected · sink ready

    ▸ streaming · lag 38ms · 300K+ events/sec

    300K+
    Events per second
    <1s
    End-to-end latency
    ~5MB
    Memory per worker
    HA
    High Availability

    From WAL to warehouse in under a second

    One Rust binary, sub-second latency, ~5 MB memory. Replicate PostgreSQL to any modern sink without overhead.

    We ship fewer connectors, but every one is built to survive production workloads.

    Sources

    PostgreSQL

    Primary source

    PostgreSQL 10+ via pgoutput — no polling, no triggers

    Native WAL tailing tracks every INSERT, UPDATE, DELETE with zero schema changes.

    Shipping 2026

    20+ more sources

    IBM DB2 · Oracle · MySQL · SQL Server · MongoDB · and more

    Soon

    ez-cdc engine

    300K+events / sec
    <1slatency
    Single Binary · No JVM

    Destinations

    StarRocks

    Real-time dashboards, sub-second sync

    PostgreSQL

    Reporting and HA replica

    Snowflake

    Near-real-time cloud warehouse

    Shipping 2026

    MongoDB

    Document sync for app reads

    Soon

    Amazon S3

    Data lake and cold archive

    Soon

    Ship your first pipeline in one command

    A CLI that does the work for you, a local web wizard when you want a GUI, and zero-config provisioning so you never touch a replication slot by hand.

    CLI — first-class experience

    ez-cdc quickstart · ez-cdc verify · ez-cdc up / down / logs / status. Interactive terminal dashboard shows stage, lag, throughput, and source→target row counts in real time. Shell completion via --install-completion.

    Web UI wizard at localhost:8080

    Build with --features http-api for a local web frontend. Test connections, discover tables, start replication in one click. No config files, no YAML edits — just a browser tab.

    Zero config — auto-provisioned

    ez-cdc creates publications, replication slots, sink tables, and audit columns automatically. No DBA approval, no schema migrations, no manual slot management. Run the command, watch it work.

    Two ways to run ez-cdc. Both free.

    Self-host the open-source engine, or click-deploy into your own cloud. Same Rust engine either way — you choose how much you want to manage.

    Self-host dbmazz

    The Rust engine, yours to run. Free forever.

    • Single binary — no JVM, Kafka, or ZooKeeper
    • CLI + interactive terminal dashboard
    • Local web UI at localhost:8080
    • Elastic License 2.0 — inspect every line
    • Community support
    View on GitHub

    BYOC — Production in your cloud

    Click-deploy into your AWS or GCP. Free.

    • Click-deploy into your AWS or GCP account
    • HA + autoscaling + load balancing included
    • Your data never leaves your cloud account
    • Same engine, same observability stack
    • Zero infra work on your end
    Talk to Founders

    Both deployment modes are free. We charge only for support (scaled to your organization) and optional PrivateLink connectivity on BYOC.

    Enterprise-grade BYOC, fully in your cloud

    BYOC deploys ez-cdc into your own AWS, GCP, or on-premise cloud with high availability, auto-scaling, load balancing, and optional PrivateLink connectivity built in. Nothing leaves your account — the engine, the control plane, and every row of data stay within your perimeter. This is the enterprise path for teams where security and compliance rule everything.

    High Availability

    Multi-AZ by default. Worker failures don't interrupt replication.

    Auto-scaling & Load Balancing

    Workers scale with throughput. Load balanced across your cloud.

    PrivateLink Connectivity

    Private network endpoints for teams that cannot route through public internet.

    Data Sovereignty

    Every byte stays in your cloud account. Nothing crosses into ours.

    Engineered for performance

    Six performance-critical design choices power ez-cdc: 300,000+ events/sec at ~5 MB RSS, sub-second latency, no JVM overhead.

    Every component optimized from the ground up. No compromises.

    100% Rust

    No JVM overhead. Pure systems programming for maximum performance and memory safety.

    Single Binary

    One Rust binary, no JVM, no Kafka, no ZooKeeper. Built on PostgreSQL's `pgoutput` logical replication protocol with LSN checkpointing.

    Sub-Second Latency

    Under 1 second end-to-end, from WAL write to sink ACK. Flink CDC concurrent snapshot algorithm for consistent initial loads.

    ~5 MB Memory

    Each Rust worker runs in about 5 MB RSS and under 2% CPU — a single binary with no JVM, Kafka, or ZooKeeper.

    Zero config

    Auto-creates publications, replication slots, sink tables, and audit columns. No YAML files, no schema migrations, no DBA approval.

    Observable by default

    Prometheus `/metrics/prometheus`, gRPC reflection API, HTTP control plane (`/pause`, `/resume`, `/drain-stop`), live terminal dashboard. Drop into your stack.

    Cloud-native architecture

    Stateless Rust workers process WAL events via PostgreSQL logical replication, delivering to analytics sinks with sub-second end-to-end latency.

    Deploys natively to AWS, GCP, or on-premise. Your cloud, your credits, your control.

    Compute

    Stateless CDC workers

    EC2 · GCE · on-prem VMs

    Object storage

    Checkpoints & buffering

    S3 · GCS · any S3-compatible

    Analytics sink

    Your destination of choice

    StarRocks · Snowflake · BigQuery · and more

    Observability

    Metrics, logs, traces

    Prometheus · CloudWatch · Stackdriver

    Secrets

    Credential management

    AWS Secrets Manager · GCP Secret Manager · Vault

    Access control

    Least-privilege by default

    IAM · workload identity · RBAC

    Architecture of ez-cdc: stateless Rust workers process PostgreSQL WAL events and deliver to any analytics sink across AWS, GCP, or on-premise clouds.

    Control Plane

    Centralized orchestration and configuration management for your CDC pipelines.

    Worker Agents

    Stateless, horizontally scalable workers that process change events in parallel.

    Multi-tenant Ready

    Designed from day one to support SaaS deployments with complete isolation.

    How we compare

    ez-cdc is a single Rust binary, free to use, with sub-second latency at ~5 MB RSS per worker. No cluster to maintain like Debezium. No volume bill like Fivetran.

    Built from scratch for performance, not compatibility layers.

    Featureez-cdcDebeziumFivetran
    End-to-end Latency< 1 secondSeconds5-15 minutes
    PricingFreeFree (OSS)Volume-based ($$)
    Memory per Worker~5 MB RSS500+ MB (JVM)N/A (managed)
    Throughput300,000+ events/sec~250 events/secVaries
    Deployment ModelSingle Rust binaryKafka + ZooKeeper clusterManaged SaaS
    Ops OverheadNear-zeroHigh (cluster ops)Zero (managed)
    LanguageRustJava (JVM)Proprietary
    Self-hosted Option
    Open Source CoreELv2Apache 2.0
    12×

    faster than Debezium and Fivetran on equivalent workloads

    Built by two brothers

    Built by two brothers who fought CDC pain on every data team they joined.

    Nahuel Mazzitelli

    Co-founder

    Co-founder. Former Head of Data at fintech and banking, built multiple data platforms. Started ez-cdc after hitting the same CDC pain point on every team.

    Darío Mazzitelli

    Co-founder

    Co-founder. AI and software engineering lead. Builds the Rust engine and the agent tooling behind ez-cdc.

    Frequently asked questions

    Technical answers for technical teams.

    Ready to 10× your CDC?

    Join technical teams who are tired of slow, resource-hungry replication. Get early access to the private beta.

    Talk to Founders

    Limited spots available. We're working closely with design partners.