Your IBM i system is the foundation of your business, and you need 24/7 availability. Any downtime can lead to business disruptions, lost revenue and unhappy customers. Rocket iCluster high-availability/disaster recovery (HA/DR) solutions help ensure uninterrupted operation for your IBM i applications, providing continuous access by monitoring, identifying and self-correcting replication problems.
Rocket iCluster reduces downtime related to unexpected IBM i system interruptions, with real-time, fault-tolerant, object-level replication. In the event of an outage, you can bring a “warm” mirror of a clustered IBM i system into service within minutes.
iCluster ensures a high-availability environment by giving business applications concurrent access to both master and replicated data. This set-up allows you to offload critical business tasks such as running reports and queries as well as ETL, EDI, and web tasks from your secondary system without affecting primary system performance.
Proactive issue notifications and self-correction for replication problems help your team identify and address potential issues before they affect the system availability or recovery solution. The administrative capabilities in iCluster help staff easily configure and perform operations such as starting, ending and switching replication groups, saving time and removing some of the risk associated with manual processes.
Rocket iCluster helps you address unexpected events by continuously replicating a recovery point to a warm standby server, which can be deployed within minutes should the need arise. iCluster mirrors changes, additions, or deletions automatically in near real-time to the replicated system—with no need for manual intervention. It ensures the highest fidelity from production to secondary instances, with always-on sync-check and byte-by-byte verification to confirm accuracy. iCluster also analyzes IBM i journal data and displays replication status in terms that are clear and easy to understand—your team can say goodbye to tedious manual reporting and monitoring processes.
Whether they’re located in the same data center or thousands of miles away, iCluster helps reduce the risk of a system outage or a regional disruption of services. Automated monitoring and alert services allow IT professionals to focus on strategic initiatives instead of constantly monitoring for replication health interruptions. High-speed operational switching ensures that outages are minimized and business resumption can be established well within your set Recovery Time Objectives.
Rocket iCluster is a cost-effective way to ensure business continuity and high availability of your critical applications and data. It’s easy to buy, easy to configure, and easy to use. You get everything you need for your IBM i environment, with a single installer and no modules or add-ons.
iCluster is licensed on a per-server basis, rather than a per-LPAR or CPW rating basis that’s common with other solutions (which can cost as much as 5x more). There are no added charges for transfers to new hardware within the same processor group, and you never have to worry about license audits or predatory upgrade pricing.
An intuitive user interface simplifies monitoring and troubleshooting, making it easy for any member of your IT team to run and operate iCluster with confidence. You can designate any server as the master (or metadata owner) for easy, fast configuration with updates to all other servers performed automatically. Instead of having to set up a special administration server on another platform or perform multiple configurations on multiple servers for switchover, the distribution and fidelity of the configuration is maintained within iCluster.
The automatic configuration command, together with iCluster’s ability to configure once on the source and then push it out to the backup system, means your business is up and running much faster, with improved availability.
With superior high-speed replication capabilities, iCluster automatically scales to handle the size of your business and its transactional volumes. High-speed replication and compression minimizes the bandwidth and time required for re-synchronizations after lengthy outages or switchovers.
Transferring, mirroring, or replicating from production servers to backups needs to happen in near real-time. However, what good is speed if the replication is inaccurate? iCluster ensures that everything is logically synchronized byte-by-byte between your source and target, giving you unsurpassed accuracy.
Each new release of iCluster includes features and enhancements to help improve performance. iCluster 8.2 expands existing parallel processing to separate, discrete functions that go beyond single-thread processing. Controls have been expanded to allow up to 50 parallel processes, speeding up activities such as analysis coverage reports, object count reports, and setting initial marked position for startup. This also translates to faster switchovers during the Change Role execution.
Support for replicating Db2 system-period temporal tables
iCluster 8.2 fully supports replicating IBM Db2® system-period temporal tables and the associated history tables. System-period temporal tables were introduced in Db2 for i in the IBM i V7R3 release.
New group type *HADR for high availability and disaster recovery
For some critical business requirements, replication to a single target is not sufficient. More customers are now replicating to both a High Availability server as well as a remote Disaster Recovery server, increasing switchover complexity and putting more demands on the IT team. To help simplify three-way switching processes, Rocket iCluster v8.2 has introduced *HADR replication group types, we have added a new group type called *HADR for this release. This is a replication group with three nodes–a primary node, an HA backup node ,and a DR backup node.
- Before roleswitch, the group replicates from the primary node to the HA backup node and to the DR backup node.
- After roleswitch, the DR node will retain the same role, but the primary node and the HA backup node will switch their roles. The group will replicate from the new primary node (formerly the HA backup node) to the new HA backup node (the previous primary node) and the DR backup node.
- This provides additional structure, automation, less complexity, and less risk when performing a switch between the production server and the HA server while invoking a switch of origin to the DR server
Parallel processing for better performance
The following commands and processes can now do their work using parallel processes:
- *CHECKSUM sync check
- DMMRKPOS for native objects (DMMRKPOS and DMSETPOS also use parallel processing for BSF. This support was added in iCluster 8.1.)
Sync check enhancements
- Detailed attributes for printer device descriptions
- SQL object detailed attributes
- User profile individual attributes: initial program, job description, message queue, output queue
- Initial library list for *JOBD
- Sync check reports can be sent to email addresses
- *CHECKSUM sync check can use parallel processing for performance
- Replication support for SQL sequences as an SQL object type
- Replication of journaling attributes for files (PF-DTA and BSF)
- Options *MEDIUM and *HIGH for object compression in the DMSETSVAL command
- Refresh of SQL triggers with CREATE TRIGGER statements longer than 32K bytes
- Automatic creation of a second remote journal for a local journal if necessary
- “Change object auditing” parameter added to DMCHGROLE and DMSTRSWO
- Cascaded MQ Series replication