Design. Move. Run. How OSStend takes over your data stack
OSStend engineers design your open-source databases and messaging around your workload, move you in with zero downtime and run them 24/7 in our cloud, yours or your data center.
45 minutes with an engineer: we review your workload and sketch the design.
- ✓Operated by AceMQ
- ✓160+ customers
- ✓10+ years
- ✓Enterprise only
OSStend is managed open source, tailored to your workload. An engineer reviews the workload and writes the cluster design, the team runs the migration and owns the cutover, and named engineers run the result 24/7 in our cloud, your cloud account or your data center, on one contract.
- PostgreSQL
- MySQL
- MongoDB
- Redis
- Valkey
- OpenSearch
- Kafka
- RabbitMQ
- AWS
- Google Cloud
- Azure
- Kubernetes
- OpenShift
- Prometheus
- Grafana
- Datadog
- Dynatrace
- Splunk
Design. Move. Run.
Every engagement follows the same three steps. An engineer designs the cluster and you sign it off before anything is built, the team moves you in with both sides live, and named engineers run it from then on.
Design
An engineer reviews your workload and writes the cluster design: size, topology, configs, plugins and versions. You sign it off before anything is built.
Move
OSStend runs the migration and owns the cutover, with both sides live until you switch. Zero downtime is the plan.
Run
Named engineers run it 24/7: upgrades on your calendar, backups, scaling and incident response, with an SLA written into your contract.
Both sides stay live through cutover. For RabbitMQ the method is federation, shovel and proxies (see the Amazon MQ to RabbitMQ guide); for databases, replication (see the RDS to PostgreSQL guide); for Kafka, mirroring (see the Confluent to Kafka guide). Traffic switches only when the new side is verified, and every move is listed in the migration guides. The broker path is described on RabbitMQ hosting, and the response and uptime terms for the Run step are on service levels. For databases you already run, see DBA services for databases you already run.
What we tailor
Eight things change from one workload to the next. Each is set in the design step, with an engineer, instead of being picked from a fixed plan menu.
| Area | What the design sets |
|---|---|
| Size and shape | Node count, instance size, storage and partition plan, sized to measured load. |
| Topology | Multi-zone, multi-region, hybrid links between cloud and data center, federation and shovel between sites. |
| Configs and plugins | The broker, database and OS settings the workload needs, plus the plugins and extensions it depends on. |
| Versions and upgrades | Long-term support builds (AceMQ LTS for RabbitMQ, with monthly CVE patching). Upgrade timing is set around the customer's change calendar. |
| Where it runs | Our cloud, the customer's AWS, Azure or Google Cloud account, their data center, Kubernetes or OpenShift, air-gapped sites, FIPS and Windows hosts. See deployment options. |
| Network and security | Private networking in the customer's account, their identity provider, their key management, their audit trail. |
| Monitoring | Metrics and alerts forwarded to the customer's Datadog, Dynatrace, Splunk or Prometheus, not a separate console. |
| Contract | One contract and one on-call team across every engine. |
The table lists what a design covers, not fixed limits. The values for your cluster are the ones in the design you sign off, and that design is what OSStend runs.
Where turnkey services stop
Turnkey services are built for the average workload, and their own documentation lists the edges. Each card quotes what the AWS documentation says, links the page and shows the date it was checked, then states what OSStend does instead.
| Area and managed service | Documented limit | What OSStend does instead | Source |
|---|---|---|---|
| Cluster sizeAmazon MQ for RabbitMQ | A cluster is three nodes, one per availability zone, or a single instance. | What OSStend does insteadThe cluster is sized to your measured load in the design step. | AWS docs, checked 30 Oct 2026 ↗ |
| SettingsAmazon MQ for RabbitMQ | AWS lists five broker values you can change: consumer timeout, heartbeat, operator policy changes, relaxed quorum queue redeclaration checks and HTTP security headers. | What OSStend does insteadEngineers hold full configuration access and change settings when the workload needs it. | AWS docs, checked 30 Oct 2026 ↗ |
| PluginsAmazon MQ for RabbitMQ | AWS documents a list of 11 supported plugins. Static shovels are not supported, a shovel cannot target a private broker, and the delayed-message exchange and stream plugins are not on the list. | What OSStend does insteadThe plugins the workload needs, and federation or shovel to any site. | AWS docs, checked 9 Nov 2026 ↗ |
| Upgrades on their scheduleAmazon MQ for RabbitMQ | AWS upgrades brokers to the latest patch version in the maintenance window, and moves a broker to the next minor version when its current minor version reaches end of support. | What OSStend does insteadLTS builds with monthly CVE patching, and upgrades planned around your change calendar. | AWS docs, checked 9 Nov 2026 ↗ |
| Connection ceilingsAmazon MQ for RabbitMQ | Per-node connection, channel and queue limits cannot be overridden, so a higher limit means a larger instance. | What OSStend does insteadLimits come from the cluster design, not from an instance-size tier. | AWS docs, checked 30 Oct 2026 ↗ |
| Kafka cluster sizeAmazon MSK | The documented maximum is 30 brokers per cluster on ZooKeeper and 60 on KRaft, and above the partition maximum per broker the configuration cannot be changed. | What OSStend does insteadBroker count and partition plan designed for your estate. | AWS docs, checked 30 Oct 2026 ↗ |
| Admin accessAmazon ElastiCache | On clusters running Redis OSS, CONFIG, SAVE, BGSAVE, REPLICAOF, DEBUG and other admin commands are unavailable. | What OSStend does insteadFull Redis or Valkey configuration, run by OSStend engineers. | AWS docs, checked 9 Nov 2026 ↗ |
Amazon MQ, MSK, ElastiCache and RDS are AWS services, so each runs on AWS only. OSStend runs the same engines in AWS, Azure, Google Cloud and your data center on one contract. These limits describe plan menus on one provider's services, not every provider. The Amazon MQ case is written up in Amazon MQ alternative, and every limit above is listed with its source in the managed service limits index. Each limit shows the date it was checked.
Who this is for, and who it is not for
OSStend is for enterprises whose workloads do not fit a turnkey plan. If your workload fits a standard plan, a turnkey service will cost you less.
Contact us if
- ✓Your clusters are bigger than a plan allows.
- ✓You need custom configs and plugins.
- ✓You run on your own upgrade calendar and change windows.
- ✓You have on-prem or air-gapped sites.
- ✓You run in more than one cloud.
Not a fit if
- ✕Your workload fits a standard plan. A turnkey service is cheaper.
- ✕You want a free tier or to sign up with a card.
- ✕You want a self-serve console and no conversation first.
Customer outcomes describes the workloads OSStend runs, and the guides cover the decisions behind them.
Questions about the process
Most buyers want three things settled before a first call: whether OSStend can take over what already runs, how a migration avoids downtime, and where the service can run. The answers are below.
Can OSStend take over a cluster that is already running?
Yes. The engagement starts with a review of the current setup. Operations are then taken over in place, or the workload is moved to a new cluster with zero downtime.
How is downtime avoided during a migration?
Both sides stay live through cutover. RabbitMQ moves use federation, shovel and proxies, databases use replication and Kafka uses mirroring. Traffic switches only when the new side is verified.
Can it run in our own cloud account or data center?
Yes. The service is fully managed in OSStend's cloud or in your AWS, Azure or Google Cloud account, or co-managed in your data center, Kubernetes or OpenShift, including air-gapped sites. See deployment options.
What happens in the first conversation?
An engineer spends 45 minutes reviewing the workload and sketches the design. Book it through Talk to an engineer.
Related pages
- Migration guideAmazon MQ to RabbitMQ migration without downtimeAmazon MQ to RabbitMQ migration without downtime: the federation, shovel and blue-green methods RabbitMQ documents, Amazon MQ's plugin rules and cutover.
- Migration guideMigrate Postgres off Amazon RDS, step by stepMigrate Postgres off Amazon RDS with logical replication and pg_dump: what to switch on, what replication skips, how cutover works and where the pause falls.
- Migration guideConfluent to Kafka migration with MirrorMaker 2Kafka migration from Confluent to Apache Kafka with MirrorMaker 2: topics, consumer offsets, schemas and connectors, both clusters live until cutover.
- OverviewZero-downtime migration, owned by our engineersZero-downtime migration guides for Amazon MQ, RDS, Redis, Atlas, Oracle, MySQL, Confluent and Elasticsearch, using each engine's documented method.
- RabbitMQRabbitMQ hosting in your cloud or on premiseRabbitMQ hosting for clusters a plan cannot hold: federation and shovel, the plugins you need and upgrades on your change calendar, run 24/7 by engineers.
- Service levelsService levels: two tiers, a written SLAOSStend service levels: fully managed and co-managed tiers, what each one covers, who is on call and how incidents are handled, with a written SLA for each.
Talk to an engineer, not a ticket queue
45 minutes with someone who runs these systems. We map your databases and brokers, the zero-downtime cutover path and which tier fits.
- ✓A design sketch for your workload, not a plan menu
- ✓A zero-downtime cutover path for what you run today
- ✓A price for that design, after the review