Linux & Cloud

Explore practical strategies for building resilient Adabas deployments in the cloud using multi-zone architectures, disaster recovery planning, and built-in security.

Adabas has long delivered high availability and disaster recovery for mission-critical systems, and those proven capabilities work on any infrastructure you choose, including the public cloud, with no changes to your Natural applications.

Wherever Adabas runs, resilience is a design decision. Many customers run Adabas very successfully on the mainframe or in their own data centers, and that remains a fully supported choice. But if the cloud is part of your strategy, or you are simply evaluating it, you should know exactly how Adabas keeps your core systems running when something fails.

Start with two numbers: RPO and RTO

High availability (HA) and disaster recovery (DR) solve different problems, so design them separately.

High availability keeps the database running through local failures, such as a crashed VM, a failed disk or a lost availability zone. The goal is seconds of disruption and zero data loss.
Disaster recovery restores service after a site- or region-wide event. The goal is a defined recovery within minutes or hours.

For each application, agree on a Recovery Point Objective (RPO), how much data you can afford to lose, and a Recovery Time Objective (RTO), how long you can afford to be down. These two numbers drive every architecture and cost decision that follows.

Adabas Cluster for Linux: HA built into the product

Adabas Cluster for Linux gives Adabas a native HA option that requires no application changes. You enable it through configuration and licensing, and your Natural programs keep running as they are.

The cluster uses a shared-nothing, primary-secondary architecture. Here is how it works:

Full copies on every node. Each node holds a complete set of data and can run on a separate host. There is no shared disk to become a single point of failure.
Synchronous replication. The cluster applies every change to the secondary nodes as part of the transaction, so a failover loses no committed data.
Automatic failover. When a failure is detected, a surviving secondary is promoted to primary. Updates are transparently routed to whichever node is primary, while read transactions can also be served by the secondary nodes.
Rolling maintenance. You can patch Adabas or the operating system one node at a time, without taking the cluster offline.

On the application side, Natural can also run in an HA setup across zones. Together, they remove single points of failure from the database up through the application layer.

If you run on the mainframe, Adabas Cluster Services and Adabas Parallel Services continue to provide proven HA and scalability on z/OS. Resilience is not something you give up by staying where you are.

A reference architecture

Adabas HA and DR reference architecture · 3 zones, 1 DR region

Adabas Cluster keeps a synchronous copy in every availability zone, so losing a zone promotes a secondary with no data loss. The DR region receives backups and asynchronous replication, and it is scaled up only when you need it.

Cloud features that strengthen Adabas

Adabas Cluster provides HA up to the level your underlying infrastructure supports. In the cloud, that infrastructure is very strong:

Availability zones. Place cluster nodes in separate zones. If a zone fails, a secondary in a surviving zone takes over.
Zone-redundant storage and snapshots. Pair Adabas backups and protection logs with durable, replicated storage for fast, point-in-time recovery.
Cross-region DR. Replicate backups, or use asynchronous replication with Adabas Event Replicator, to a second region you can activate after a regional outage.
Containers and Kubernetes. Adabas & Natural run on Docker and Kubernetes, so failed containers restart automatically and environments are rebuilt consistently from code.
Infrastructure as code. Define the whole HA and DR estate in Terraform or similar tools, so you can test failover regularly and rebuild a region on demand.
Administration. Adabas Manager provides browser-based administration of your Adabas databases.

The same designs work on any major cloud, or across your own data centers. You are not locked into one provider.

Security by design

Resilience means little without security, and Adabas covers both layers:

Encryption at rest. Adabas Encryption for Linux protects database files, on premises and in the cloud.
Encryption in transit. ADATCP(S) secures client traffic and TLS encrypts communication between cluster nodes.
Access control. Natural Security controls who can run which programs and reach which data, and Adabas role-based access control (RBAC) lets you grant and revoke database permissions.
Network isolation. Private subnets, security groups and private endpoints keep the database off the public internet.
Key management. Adabas Encryption in addition to local configuration also works with external key management systems, including Azure key vault, AWS Key Management Service (KMS) and HashiCorp Vault.

These controls help regulated customers meet requirements such as GDPR, DORA and NIS2, wherever the workload runs.

Cost: pay for the resilience you need

Not every application needs the same protection. Match each workload to a tier based on its RPO and RTO:

A few habits keep costs in check:

Keep the DR region small and scale it up only when you fail over.
Use reserved or committed capacity for the always-on primary nodes.
Move older backups to cheaper storage classes automatically.

Because the cluster is a configuration and licensing option, you can start with one tier and move up later without redesigning the application.

Proven wherever you run

Adabas has always been trusted for mission-critical transactions. That same proven reliability applies on Linux, containers and every major cloud, with native clustering, automatic failover, encryption and flexible DR options. Whether you stay on the mainframe, run in your own data center, or plan a move to the cloud, the choice is yours. If you would like to review the HA and DR design for your current or future environment, our architects are happy to help.