If anything, the 2017 British Airways IT systems failure emphasized the need for enterprises to pay more attention to data protection and business continuity. In the past, there were redundant paper systems to resurrect information. As a result, outages or minor data loss was inconvenient rather than tragic. Today, however, businesses must have some kind of high availability or disaster recovery capability in their infrastructure. This is particularly essential for enterprises running on IBM i, because IBM i systems run high-volume transactional workloads. As a result, if your IBM i transactions become lost or corrupted and your system is unavailable, then your business operations essentially halt. That is why you must consider high availability (HA) and data recovery (DR) in your IBM i application modernization journey.
In our fast-paced economy, enterprises can’t afford any unscheduled outages. Customers expect 24/7 system availability. In many cases, even overnight or weekend outages are unacceptable. As a result, many IBM i businesses struggle with the impact of planned and unplanned downtime on their stakeholders, customers, suppliers, and trading partners. Fortunately, with the right IBM i application modernization strategy, you can implement great HA and DR capabilities. Here are 11 tips to guide you through high availability and disaster recovery.
1. How Can You Gauge Business Requirements?
Implementing high availability and data recovery is a considerable investment. Doing it right involves establishing clear business requirements before moving to a solution. The process begins with gauging your current requirements and having all your business units sign off on what’s critical to their operation. For example, if a company has many orders coming in via fax, their fax machine is a critical piece of infrastructure.
Simply put, there’s no point in a disaster recovery solution for your core application if other critical elements of your stack remain unprotected. You need a service that covers all of your bases.
2. Should You Define What Solutions You’re Implementing and What Risk You’re Mitigating?
Like any IBM i modernization strategy, application availability and data recovery begin with defining both the risks you intend to mitigate and the solution that achieves those goals. This includes your recovery time objective or how quickly you can have systems back up and running if you suffer an outage either planned or unplanned. After that comes determining how much data you can afford to lose. These factors will guide you to the right solution. According to IBM documentation, the following parameters determine the data recovery solution best for you:
- Budget
- Uptime requirement
- Outage coverage
- Recovery time objective (RTO)
- Recovery point objective (RPO)
- Resilience requirements
- Automated failover and switchover
- Distance requirements
- Number of backup systems
- Access to a secondary copy of the data
- System performance
- Data resilience method comparison
3. Should You Monitor the Capabilities of Your Solution?
To maintain the efficacy and performance of your solution, you need to monitor these factors in terms of your business requirements. This also includes having the broad level of experience required to conduct disaster exercises.
Explore more information on how your organization can utilize modernization solutions in this article about IBM AS 400 iSeries.
4. Should You Stress Test Your High Availability and Data Recovery Process?
Once you’re clear on what solutions you’re implementing and what risks you’re mitigating, you need to practice!!! Stress test your data recovery process. If your data recovery process involves multiple people, then properly train and test them all. You should also plan for different scenarios, with different members of staff available. Attacks don’t happen when it is convenient. If an attack or failure happens on a day like Christmas, for example, you need to be properly prepared to deal with it using limited resources.
5. What About Reviewing Your Tests?
In addition to testing and practice, review the tests you’ve done and be disciplined about improving your processes. Having processes in place is the key here. Create processes that prioritize what you’ve learned from each test, and incorporate those lessons in subsequent tests.
6. Should You Document Your Plan?
Documentation is essential to guide you in the event of a real-life disaster. In the process of running numerous tests, many forget that a disaster could mean that you no longer have a building. So, make sure it’s accessible regardless of your building or the people in it.
7. Should You Have Multiple People That Can Run the Core Function?
Again, this boils down to planning for different scenarios. Your outcomes can only be as successful as your preparation. Train enough personnel that each shift has key people present.
8. How Should You Ensure Fluid Communication Within the Business?
It’s great to have all your networks integrated so your different systems communicate and share data. Even better, however, is clear communication between people within your business. Unfortunately, communication about DR plans rarely extends beyond key people in IT. In the event of a system failure, everyone in your business needs to know what’s expected of them. In the middle of a DR event, ensure you communicate with staff as well as external stakeholders. If people know what’s going on, then they remain calm. Good communication reduces anxiety.
9. Do You Need to Have a Back-to-Normal Plan?
During HA and DR processes many businesses get caught up in technical processes and neglect back-to-normal, or business continuity planning. After all, business continuity is the point of DR, so plan exactly how you will revert to normal operations. It doesn’t have to be overly complex, but you must consider time frames and costs.
10. Do You Have A Clear Authority On Who’s Allowed to Initiate the Data Recovery Plan?
Don’t lose time during a major outage with key staff who feel unauthorized to declare a disaster. Ensure there is always someone on deck with the authority to make that call quickly.
11. Should You Develop a Data Backup Plan?
According to ready.gov, you can develop a data backup plan by “Identifying data on network servers, desktop computers, laptop computers and wireless devices that needs to be backed up along with other hard copy records and information. The plan should include regularly scheduled backups from wireless devices, laptops, and desktop computers to a network server. Data on the server can then be backed up. Backing up hard copy vital records can be accomplished by scanning paper records into digital formats and allowing them to be backed up along with other digital data.”
Are You Ready to Implement High Availability and Disaster Recovery?
The tips discussed in this article are a loose guide to implementing HA and DR. However, you must set yourself up for success. Recovering data silos can be very challenging, and IBM i applications are often monoliths. This means that they produce isolated data that is difficult to integrate with existing systems. Fortunately, IBM i application modernization offers a way out. With tools like aXes and LANSA’s Cloud Strategy, you can better position yourself for success.
