Living in an Active World: Why You Need an Active-Active High Availability Cluster

Posted by VIOLIN SYSTEMS on May 4, 2020 7:00:00 PM

Can you imagine a world in which you only use half your brain’s processing power or half the chocolate chips in a chocolate chip cookie recipe or a world where you only fill half your gas tank— even on long road trips?

While these ideas might seem crazy (after all, who would want a cookie with half the chocolate?), this is how some high availability digital storage marketing companies operate. The reasoning? They tell their customers to reserve half of their storage potential and save it only for use if the first half stops working.

If you’re the CIO for a company and managing decisions about how your transformation to a totally digital storage solution should happen, you’re not alone. As you evaluate your options, you should consider the differences between active-passive storage systems, mutual active-active systems, and active-active systems, and why we think active/active storage is a much better option for you.

High-Availability Server Clusters

When we talk about high-availability cluster storage, what are we referring to? High-availability clusters are fairly standard in enterprise storage systems; it’s a cluster of storage hosts or controllers that act like one single system to provide continuous service.

The theory behind many high-availability clusters is that to prevent service outages, you prevent component failures by having at least two of everything. This means you purchase two storage controllers to act as the architecture needed to run the critical applications of your business. If something goes wrong, the virtual machines on your storage system can switch from a failed server to a functioning one (also referred to as a “failover”), with minimal delay in service.

This is a software-designed storage approach: It means that you are creating storage systems that uncouple your storage resources from the hardware platform. It gives you greater flexibility and efficiency, especially when issues occur.

There are three available options when considering what kind of storage controller to purchase:

  • Active-passive controllers
  • Mutually active-active controllers
  • Active-active controllers

Active-Passive Design

In an active-passive design storage system, all the critical operations of your business happen in one of two controllers. Essentially, you have an active connection to one controller where all the data is kept, and a secondary, passive connection on standby that leads to a second, empty controller.

In the event of a “pathway failure,” like a network connection issue or an HPA failure, everything has to failover to your second cluster, which takes time. During this time, your team or your clients may be without access to vital data.

In this active-passive scenario, you are paying for this second half of your storage solution to do nothing the majority of the time. This means that you are limited to only 50 percent of the performance you are paying for. You’ve purchased two controllers, but you only are receiving the benefits of having one controller. It’s like buying a second house in case the lights go out in the first house!

Mutually Active-Active Design

A mutually active-active storage controller system isn’t much different. In this instance, you store half of your data in one storage controller and the other half in the second controller. The problem here? If there’s a pathway failure in one of your storage controllers, you still have delay time while waiting for that first set of data to failover and transfer to the second controller.

In a mutually-active, active design, both connection pathways to access your load are active, but they are only active for the half of the load stored in that controller. You still have the same lag in performance waiting for the load to failover to the other controller when a pathway fails, just like you see in an active-passive controller system.

Active-Active Design

How can you prevent this failover delay in storage retrieval? Enter the active-active controller design. In an active-active controller system, both controllers are available to handle your critical operations. You can store data in both controllers, which means you double your performance and your storage space, for the same price as an active-passive controller.

And when an active-active controller fails? If the first controller fails, you still have access to the second controller. You have half of your normal performance capacity, but this is the same amount of performance capacity you would have with an active-passive controller system. Performance under failure is the same for both systems.

If active-passive controllers can be equated to paying for two houses but only being able to use one, an active-active controller means that you pay for two houses and you get to use two houses. In the event something happens to your first house? You still have somewhere to stay.


When you want to scale up with an active-active controller system, all you have to do is add another controller. As long as your current setup will support additional controllers to be added and active at the same time, scaling up is as easy as adding another controller.

With a mutually active-active controller system or an active-passive system? Scaling is much harder and may make performance worse rather than improve things. The only solution here is to replace your controllers with larger, more powerful controllers because you can’t add to the existing ones.

The Benefits of All-Flash Storage

Violin Systems is a strong believer in the power of all-flash storage that’s available with active-active systems. Why is that? As your team becomes increasingly mobile (and as many of us work from home or contend with working remotely) digital transformation has to be fluid, and it has to work, no matter where you’re accessing the information.

How can all-flash storage help with this?

  • Increased performance, agility, and ability to innovate and change (without being tied to hardware)
  • Lower cost of ownership
  • Increased reliability and longer mean time to failure

At Violin Systems, we believe you should get what you pay for. If you’re paying for two controllers, we think you should be able to use both. It’s as simple as that. This is why we offer active-active storage controller systems; because we think it’s the best combination of exceptional value and high-end performance.

No matter what kind of storage solution you’re seeing we are here to keep your company up and running. When you work with us, whether your team is on-site or working safely from home, you can trust your systems to keep up with the demand, even when there’s a failover.

Want to learn more? Contact us today!