Hardware High Availability (HA) and logical replication are both smart choices when deciding how to…
With every High Availability (HA) solution your company considers, you need to feel confident that your data is protected at the point of failure on the production server. For this to happen, the target system, or target SAN, has to be readily available and operating whenever the production environment is up and running. Target systems can be impacted by maintenance operations, as well as unplanned system outages on the target system. This is why it is important that you choose an HA and disaster recovery (DR) solution that will protect your data—but more importantly recover it.
Options for Target-Side Solutions
Logical replication allows for real-time read access to the target system’s data. If logical replication is used between independent auxiliary storage pools (IASP), the target-side IASP is varied on and available at all times. IASP a is switchable technology where local journaling of all objects takes places to support recovery after a failure.
Commonly, the active, target-side data is used for queries and offline saves to tape. This can benefit your company because when offline tape saves are operating, the logical replication will be stopped, while the remote journal operation continues to send data changes to a buffer on the target system. The buffer of unapplied journal transactions are stored until the tape save operation is completed—or the save while active checkpoint is determined—the apply operations are then restarted. If your systems fail during the process the tape save operations can be stopped, and the apply jobs can then be restarted. Once all the journal entries stored in the buffer are applied, switching can then be performed.
Throughout this process, your system’s latency status is reported to ensure your business continuity is not severely impacted by the process.
Hardware HA technologies, like Geographic Mirror, Metro Mirror and Global Mirror, are based on IASPs, which can only be accessed when varied on and available. When replication is happening, the IASP copy on the target system is attached, and essentially becomes a part of the source IASP environment. When the IASP copy is attached, only the source system can access it. When the IASP copy on the target system is detached, the original and copy become two separate IASPs. They can then be varied on independently of each other. When this happens, changes made to the target-side IASP are tracked, and the IASP can be restored to its original state when it is reattached to the source system.
The changes made to the data on the source side are buffered while the IASP copy of the target is detached. During this process, the target-side changes are removed, and the source-side changes are applied.
The buffering of source data to send to the target and target-side changes, where it will be returned to its previous state, consists of memory maps. Due to this, a large amount of data changes can be buffered. If this solution is chosen, it is important for your company to define their limits for a detach operation. This can be in terms of:
- how much data can be changed on the source
- how much data can be changed between source and target
- how much time is necessary for the reattach operation to re-sync the IASP copy with the source
The Best Protection for Your Company
Planned and unplanned interruptions will happen to every company, and the key to not letting them impact your business continuity or data is being prepared. Both hardware HA and logical replication have solutions for protecting your company’s data and systems—but which is best for you? Contact Baseline Data Services today at (317) 707-3941 for guidance in determining what is the best fit for your company. We have years of experience protecting data and designing DR plans for a large array of companies. Let us help you keep your systems safe with the best HA solution and DR plan for you.