cloud optimization

If your cloud optimization strategy is once and done, you’re not actually optimizing your application. 

Here’s why. 

 This strategy is similar to watering your crops when you plant them and not adjusting for the season, the changing weather conditions, and certainly not watering for the season’s duration. If this were the case, your crops wouldn’t stand a chance. Yet, this is how most companies are approaching their cloud optimization strategy today.

Now, you’re probably wondering if this strategy is so inefficient, why is it so common? The answer is simple because it is the “easiest” method. Does this mean it is the most efficient in regards to your application performance, reliability, and spend? Absolutely not. SREs are deploying the first configuration that meets the SLO or the first configuration that didn’t fail. Not the most optimal configuration. 

Does this sound like the right way to approach your cloud?

Today, most optimization strategies configure the application for a 90th percentile use case, which in reality means the application is only right-sized for a very small percentage of the time. The application is either over-provisioned frequently, under-provisioned moderately, and only correctly provisioned/ optimized for a minimal amount of time. Who’s going to suffer for this? Your end-user. Your application is at risk for outages and sub-optimal performance resulting in an unfavorable customer experience. 

The solution to this? Implementing a cloud optimization strategy that actually tunes and right-sizes your resources.  

Intending to improve your end-user customer experience, we created Continuous Optimization. Opsani’s Continuous Cloud Optimization is an innovative solution that works by optimizing for the correct configuration at the right time. How is this done? By avoiding point in time optimization for a peak load.  Like we said earlier, a once and done strategy for optimization will simply not cut it.  

Let us paint you a picture. In a production scenario, there are changing conditions, changing loads, infrastructure changes, and code changes that all affect performance and demand. These settings aren’t going to manage and maintain themselves. Reconfiguration and re-optimization are required to improve reliability, performance and reduce cost.    

In most production environments today, where humans are configuring and optimizing the applications, the configuration promoted to production is not optimal; it’s merely the first configuration that didn’t fail and met the requirements. Which, not surprisingly, will result in ramifications in the long run. 

 Unlike other recommendation engines that simply tell what’s wrong with your application, Opsani takes actions to remediate it. Opsani’s Continuous Optimization technology produces high availability, lower cost, best performance, ensuring that you are never paying more than what’s necessary to meet your Service Level Objectives. We do this by monitoring your application and then autonomously adjusting resources and application-specific parameters to reoptimize your application for the highest performance and lowest cost. Producing the most optimal configuration that meets and/or exceeds your defined Service Level Objective for the application. Check out our last blog, Why Recommendation Engines Aren’t Optimizing Your Application, for more information on Opsani’s technology.