How to Shrink High Availability Overhead on IBM i
Guide

How to Shrink High Availability Overhead on IBM i

Foreword

FOREWORD

Over the last five years, studies have shown that high availability and disaster recovery is a top concern for the majority of organizations running IBM i. Adoption of high availability as a recovery system has increased, while reliance on tape alone is steadily declining.

There are a hundred and one ways to do data replication on IBM i. Logical, host-based replication technology predates storage, hardware-based replication and continues to be popular on IBM i, though many use PowerHA or a combination of both.

With nearly 50 percent of IBM i organizations already relying on high availability for disaster recovery, why does it continue to be a top concern?

There are two broad groups here: the have nots and the haves. Among the first half without a high availability solution, you’ll find organizations that are apprehensive about starting the project. They know they need the additional layer of recovery protection to comply with certain requirements, but they view high availability as expensive and complex—even unnecessary—and delay as long as possible. Among the second half of organizations, you’ll find many that are not very happy with or confident in their high availability solution but are scared to make a switch.

This guide zeros in on the second group, those who already have a high availability solution in place. We’ll explore the reasons why these solutions may no longer be meeting expectations and share advice on how to safely switch to something better. You’ll learn ways to:

  • Minimize the impact on system performance
  • Reduce total cost of ownership
  • Build confidence in your ability to recover

Enjoy!

How Overkill Installs Impact Performance

How Overkill Installs Impact Performance

A common complaint for some of the best-known logical replication solutions available for IBM i is that they are CPU-intensive, sometimes hogging 15 percent or more of production server resources.

For some large enterprises producing 100 million records on the system every day with very heavy data replication needs, it’s easy to be deceived into thinking that you need to sacrifice performance in exchange for real-time replication. Not necessarily.

It’s important to understand why certain solutions make such a dent in system performance. There are two main reasons: the method for finding new objects to replicate and the frequency of integrity checks.

Some IBM i high availability solutions read through QAUDJRN for new and changed objects instead of reading in intervals for user profiles and spooled files. With today’s security concerns, for example, every security event is hitting this same operating system object. Consequently, what made sense 10 or 20 years ago for determining new objects does not perform well anymore—your audit journal has millions of entries that don’t correlate to high availability or need to be replicated. This point alone has a major impact on your production system.

Integrity checking is a confidence limitation that stems from the QAUDJRN piece. If you’re worried that your high availability solution is out of sync, you’re more apt to trigger integrity checks to compare the current state of your data to a previous state in order to detect any changes. Many organizations run integrity checks every day because the performance issues mentioned above cause replication to get behind. Of course, frequent integrity checks throw additional overhead onto the system, too.

TAKEAWAY: Stay focused on the data you need to replicate. There are better, more efficient ways to grab new and changed objects. Everything else is overkill.

What Drives Up the Cost of High Availability Solutions?

What Drives Up the Cost of High Availability Solutions?

As new high availability options have entered the market, the old one-size-fits-all software model no longer makes sense. You scale your IBM i server based on your needs, so why not do the same for your HA solution?

True, high availability comes with inherent additional expenses to cover dual licenses for both source and target systems. If you’re upping your high availability game or just starting out, you may also get hit with expenses for a communications link with suitable bandwidth, additional hardware, a remote backup site, and new licenses for the instance of your business applications on the backup box.

The good news is that there are some costs that you can control.

As with any software, a higher initial purchase price leads to higher annual maintenance for technical support and software improvements, which leads to higher hardware change fees and, ultimately, a higher total cost of ownership (TCO). Switching to a solution with a lower initial purchase price will also reduce your TCO.

In addition, complex solutions often require a dedicated administrator to monitor and configure complicated software, which adds to headcount and compounds cost concerns. Switching to a solution that is simple and largely self-managing is another way to reduce your TCO.

Higher hardware costs are also associated with the higher consumption demands of complex high availability software. Having both a software and hardware vendor you can trust is critical. Some hardware vendors won’t tell you about CBU hardware and DRaaS offerings. They would rather sell you a full price duplicate server.

TAKEAWAY: All other things being equal, a lower initial purchase price will always reduce the total cost of ownership for your high availability solution.

 

How to Cut Out the Complexity in High Availability Solutions

How to Cut Out the Complexity in High Availability Solutions

What are the signs that your high availability solution is too complex? The easiest way to tell is when you have dedicated staff that must monitor and configure the software every day. Your solution should audit and monitor itself. That’s not to say that simple solutions require zero monitoring. Rather, the software itself or the vendor behind it should offer automation to help you out.

Another indicator is how long the tool takes to set up. You should be able to begin replicating in a matter of hours, not days. That said, with something as important as your irreplaceable business data, implementation services and product training are always a good idea, even for experienced IT teams.

The common theme between these two indicators is ease of use. Once it’s up and running, is your software so intuitive that you can use it without fear you will break something? Or, does your vendor insist—or even require—that you reach out to them to make any changes to your configuration? Talk about a roadblock. Your software should give you the flexibility and your vendor should allow you the freedom to manage it on your own.

For reference, Robot HA uses a straight-forward menu system that doesn’t hide its features behind complex terminology or options. Administrators who are used to other, more complex systems are often amazed at how much easier it is to replicate and role swap with Robot HA.

Maybe you don’t want your vendor to interfere with your data replication, but you don’t want to be responsible for it either. That’s okay! For total flexibility, HelpSystems partners with managed service providers (MSPs) to bring your IBM i high availability into the cloud. We can implement together and let the MSP monitor your target system and HA activities.

TAKEAWAY: If it sounds too complicated, it is. If your team insists that they need the most expensive and most complicated solution, challenge them and give HelpSystems a call.

How to Mitigate Risk When Switching High Availability Solutions

How to Mitigate Risk When Switching High Availability Solutions

Some industry information shows that 85 percent of respondents have no recovery plan or are less than 100 percent confident in their plan. The information also suggests that 20 percent plan to change HA/DR solutions or implement one for the first time in the next year.

If you’re one of the lucky few with unshakeable confidence in your existing IBM i high availability solution, congratulations! But what about the rest of you?

If you don’t have a recovery plan, the first thing to do is get one ASAP. HelpSystems recommends a layered approach of real-time replication and point-in-time backups to cover all your bases.

Secondly, if you’re not confident with your current HA solution, rip it out and replace it. HA/DR is too important a project to gamble on. Choose a trustworthy vendor with proven technology and experience supporting high availability transplants.

Hundreds of organizations rely on Robot HA for business continuity on IBM i, and HelpSystems technical support is the best in the business. Robot HA scales to accommodate IBM i shops of all sizes, and automation minimizes ongoing, manual maintenance efforts for your admins.

Perhaps best of all, Robot HA can be set up in hours, not days. The default configuration provides a great foundation and professional services can help you get up and running quickly. Plus, its simplicity is legendary, so admins at every level can manage the tool with ease.

HelpSystems understands this strategic technology and we want to make it easy for you to guarantee the availability of your irreplaceable business data. That’s why we’ve joined forces with IBM to share the development roadmap for PowerHA. Two powerful tools under one roof!

TAKEAWAY: HelpSystems has been a trusted IBM i vendor for years. We stand behind our software and we are honored that our customers stand behind us.

Case Study: Robot HA Keeps Nesco Resource Out of Harm’s Way

Case Study: Robot HA Keeps Nesco Resource Out of Harm’s Way

In the weeks leading up to Hurricane Irma in 2017, the Nesco Resource IT team heard the warnings and knew their Tampa data center was at risk. IT Director Gerald Dalton knew they couldn’t withstand any downtime. Fortunately, Dalton was ahead of the game.

Shortly after joining the company in 2010, he had recognized the need for backup capabilities in another geographic location. The company established a backup system in Winter Haven, Florida and implemented the Robot HA high availability solution for IBM i from HelpSystems.

When Hurricane Irma began making its way to Tampa, Dalton was ready. “We decided to do a full role swap to the facility in Winter Haven as a preemptive strike,” he said. “We also asked the HelpSystems staff to assist, which was a wise move. They took care of everything, and we were up and running quickly.”

Dalton’s trust in Robot HA is strong, which is important given the safety of more than one million records is at stake. “We did lose power for two days in Tampa during the hurricane, so it was a smart move on our part to do the role swap,” Dalton said. “Robot HA and the HelpSystems support team made it seamless for us. Not only did they give us great service, but they also followed up to make sure things were stable. That means a lot to us IT guys.”

Dalton is also prepared for the ongoing pace of growth the company has been experiencing. “Nesco Resource has doubled in size over the past eight years, and I expect it will double yet again in the next two to three. With Robot HA, we’re ready.”

Dalton is pleased with his HelpSystems experience. “I absolutely cannot say enough about Robot HA. No matter where I’ve installed or used it, it’s been a great product. I’m happy it’s under the HelpSystems umbrella.”

"Robot HA and the HelpSystems support team made it seamless for us. Not only did they give us great service, but they also followed up to make sure things were stable. That means a lot to us IT guys."
—Gerald Dalton, IT Director | Nesco Resource

Switch and Smile

Switch and Smile

Running the right high availability solution is more important than ever. Yet many administrators are fed up with their cumbersome, expensive tool.

Not sure if your team is ready for a switch? Here’s a quick gut check:

  • Is your system performance suffering from the impact of your current solution?
  • Are you afraid to look at your annual maintenance or latest hardware upgrade charge?
  • Are you afraid to test recoverability using your current solution?

If you said YES to any of these, it’s time to look for a new high availability solution.

Remember, a little research reflects positively on your initiative and no one outside of your own organization can force you to switch. At HelpSystems, we won’t recommend a solution if it’s not right for your business. Every time Robot HA replaces an existing solution, we want to make sure that there’s a smile on our customer’s face.

That’s why we offer a special SWITCH AND SMILE package to organizations looking to move away from their legacy HA software. We make it easy for you to run these solutions side by side in the same environment before you switch, so you can be confident you’re making the right choice.

You can learn more about our special offer and get pricing when you visit the SWITCH AND SMILE hub at www.helpsystems.com/switch-and-smile.

Related Products

Related Solutions