Apache Flink alternative — Apache Flink vs Skippr

Apache Flink is Apache-2.0 stream and batch compute. JobManager plus TaskManagers run SQL, Table API, or DataStream jobs. Current stable release is 2.3.0 (25 Jun 2026). Iceberg is not in the engine. Iceberg 1.12.0 (29 Sep 2026) ships iceberg-flink-runtime-2.3-1.12.0.jar (also 2.2, 2.1, and 1.20; Flink 2.0 was dropped). Catalogs: Hive, Hadoop, REST, Glue, JDBC, Nessie. Streaming INSERT INTO appends. INSERT OVERWRITE is batch-only. UPSERT needs Iceberg v2 plus a primary key (write.upsert.enabled or the upsert-enabled write option) and cannot run with overwrite. The Flink Iceberg sink documents exactly-once. Streaming reads start from a snapshot (streaming='true', monitor-interval). Creating tables with hidden partitioning is still listed as unsupported. CDC is a source you add: Kafka, or the separate Flink CDC YAML layer. There is no licence fee. You pay JobManagers, TaskManagers, checkpoints, and a catalog. Cloud Flink is Ververica, Confluent Cloud for Apache Flink, or AWS Managed Service for Apache Flink. This is not Iceberg ∪ WAL you query in one binary.

Skippr is an open Streamhouse you run as one Rust binary. Write skippr.yml (or the same thing in Python), discover, sync. You query live state over Iceberg ∪ WAL, and the history is ordinary Iceberg in your bucket. No JobManager. ELT into a warehouse is one path, not the product.

Feature	Skippr	Apache Flink
Deploy	Single Rust binary, your VPC or on-prem	JobManager + TaskManagers (K8s, YARN, standalone, managed Flink)
Schema	In-stream evolution, sibling columns	Flink SQL / Table schema plus Iceberg DDL
Storage	Open Iceberg on Parquet you own	Iceberg via the runtime jar. Catalog is yours
CDC	Log-based; fails closed if the dest cannot honour	A source connector or Flink CDC. Not Flink itself
Query	`skipprd query` on Iceberg ∪ WAL; history in any Iceberg engine	Flink SQL on the job. Iceberg after the sink commits
Pricing	Free self-hosted engine	Apache-2.0. Pay TaskManagers. Cloud Flink is CFU / KPU / Ververica.

What Apache Flink is good at

Stateful stream compute at cluster scale: windows, watermarks, exactly-once sinks, and the same SQL on bounded and unbounded inputs. That is the product. Iceberg as a DataStream or Table sink is why lake teams still submit a Flink job instead of a batch rewrite. If the job is "this Flink query into this table," that is a real product.

Where the job is wider

Flink is a compute cluster. It is not a Streamhouse. There is no Iceberg ∪ WAL. There is no skipprd query on in-flight rows. Postgres CDC is Flink CDC or a Kafka topic, not flink run. Iceberg overwrite will not run as a streaming SQL insert. Upsert is a v2 PK table. Hidden-partition CREATE is still missing.

The cluster is the ops. Checkpoints, TaskManager memory, catalog jars, and a runtime that must match Iceberg (2.3 / 2.2 / 2.1 / 1.20 on 1.12.0). Skippr will not replace a Flink job that already joins streams. Skippr also will not run a JobManager. Flink CDC is a separate page. Spark Structured Streaming is too.

If the job is "Flink we already staff," Flink is the narrower tool. If the job is live SQL on Iceberg you run, the cluster was never that.

Getting started

Install skipprd from the install guide, or pip install skippr. The three tabs write the same skippr.yml. Then discover, sync, and query.

YAMLCLIPython

yaml

skippr:
  workspace: analytics
  skipprd_el_storage_mode: local

pipelines:
  orders:
    data_source: data_sources.app_db
    data_sink: data_sinks.lake

data_sources:
  app_db:
    Postgres:
      host: db.internal
      user: skippr_reader
      password: ${APP_DB_PASSWORD}
      database: app
      tables: [orders, customers]
      cdc_mode: snapshot_then_cdc

data_sinks:
  lake:
    AthenaIceberg:
      warehouse: s3://my-iceberg-warehouse/
      glue_database_name: analytics
      athena_workgroup_name: primary
      athena_results_s3_bucket: my-athena-results
      region: us-east-1
    schema_sink: schema_sinks.lake_schema

schema_sinks:
  lake_schema:
    AthenaIceberg:
      warehouse: s3://my-iceberg-warehouse/
      glue_database_name: analytics
      athena_workgroup_name: primary
      athena_results_s3_bucket: my-athena-results
      region: us-east-1

bash

skipprd --workspace analytics --storage-mode local connect data-source postgres \
  --pipeline orders \
  --name app_db \
  --host db.internal \
  --user skippr_reader \
  --password '${APP_DB_PASSWORD}' \
  --database app \
  --tables orders \
  --tables customers \
  --cdc-mode snapshot_then_cdc

skipprd connect data-sink athenaiceberg \
  --pipeline orders \
  --name lake \
  --warehouse s3://my-iceberg-warehouse/ \
  --glue-database-name analytics \
  --athena-workgroup-name primary \
  --athena-results-s3-bucket my-athena-results \
  --region us-east-1

skipprd connect schema-sink athenaiceberg \
  --pipeline orders \
  --name lake_schema

python

from skippr import (
    Config,
    DataSinkAthenaIceberg,
    DataSourcePostgres,
    DataSourcePostgresSourceCdcMode,
    EnvRef,
    LocalStorage,
    Pipeline,
)

cfg = Config().workspace("analytics").storage(LocalStorage())

app_db = cfg.data_source(
    "app_db",
    DataSourcePostgres(
        host="db.internal",
        user="skippr_reader",
        password=EnvRef("APP_DB_PASSWORD"),
        database="app",
        tables=["orders", "customers"],
        cdc_mode=DataSourcePostgresSourceCdcMode.SNAPSHOTTHENCDC,
    ),
)

lake = cfg.data_sink(
    "lake",
    DataSinkAthenaIceberg(
        warehouse="s3://my-iceberg-warehouse/",
        glue_database_name="analytics",
        athena_workgroup_name="primary",
        athena_results_s3_bucket="my-athena-results",
        region="us-east-1",
    ),
    schema_sink="lake_schema",
)

cfg.pipeline("orders", Pipeline(data_source=app_db, data_sink=lake))
cfg.save("skippr.yml")

bash

export APP_DB_PASSWORD='your-postgres-password'
skipprd discover --pipeline orders
skipprd sync --pipeline orders --once
skipprd query "SELECT count(*) FROM analytics.orders"

Which to choose

Choose Flink if you want a JobManager cluster, and Iceberg as a Flink sink on that cluster is enough. Use Flink CDC when the log is the source and YAML is the job.

Choose Skippr if you want one binary, live SQL on Iceberg ∪ WAL, and files that stay yours. Skippr will not run Flink.

Install from Install. Connector and query reference live in Developer Docs. For optional onboarding in your VPC, Enterprise Jumpstart or Contact.

Apache Flink alternative — Apache Flink vs Skippr ​

What Apache Flink is good at ​

Where the job is wider ​

Getting started ​

Which to choose ​

Apache Flink alternative — Apache Flink vs Skippr

What Apache Flink is good at

Where the job is wider

Getting started

Which to choose