Solution icon
Governance - Cold Storage

Reduce your Apache Kafkastorage costs

Conduktor's Cold Storage enables data to be stored at low cost for as long as you need.

Apache Kafka storage costs can get out of control... not anymore

Cold storage lets you use much cheaper storage without changing a single thing in your applications

  • Feature icon
    Reduce storage costs

    Utilize any storage compatible with S3 Api like GCS, Minio, Nutanix, Netapp etc.

  • Feature icon
    Enable infinite retention

    No need to age out data for cost reasons, economically store your events for as long as you'd like

  • Feature icon
    Enable large messages

    Stop worrying about rearchitecting your applications to support large messages. Just use the Kafka client you're already using

  • Feature icon
    Rapid Recovery

    When a broker goes down, new brokers need to copy much less to get started

  • Feature icon
    Lower networking costs

    Previously, sending 1 gb in, storing it and having 3 different reads would mean you pay for 4gb. Not anymore...

  • Feature icon
    Compatible with multi-tenancy

    Everything is more efficient, with more Kafka clusters and lower storage requirements per machine

Whiteboard top visualWhiteboard bottom visual

How does it look?

Let's walkthrough some of the magic

  • Activate

    You're one API call away: a single change to your Kafka setup gets you started with cold storage.

    Use cold storage per topic or for your entire Kafka.
    Keep your Kafka messages anywhere you like to save costs

    OSX Dots$ curl --request POST http://conduktor-platform/london --data-raw '{ "type": "COLD_STORAGE", "topic": "jack-topic" }'
  • Send Large file

    Use any Kafka client under the sun and send whatever messages you need to.
    You won't need to worry about size limits.
    We can send 150Mib to S3, no problem

    OSX Dots$ kafka-producer-perf-test --topic jack-topic --num-records 1 --record-size 157286400 --producer-props bootstrap.servers=conduktor:9092 max.request.size=162529280 buffer.memory=162529280 --producer.config london-jack.properties 1 records sent, 0.3 records/sec (34.98 MB/sec), 3208.0 ms avg latency, 3208.0 ms max latency.
  • Confirm

    Our 150MiB message is stored in AWS S3. We won't need to worry about it anymore, and can keep it forever

    OSX Dots$ aws s3 ls s3://my-bucket --human-readable 2022-09-20 23:14:36 150.0 MiB jack-topic/1f4fbc2e-fd2a-40c1-8b27-f4ce5334785c

Reduce storage costs and remove frictionwithout lifting a finger

To see everything we can throw at your Kafka, arrange a demo