Every company on earth will claim that they want their cloud and mobile apps to run as quickly and as cost-effectively as possible.
However, almost none of them are carrying out real, impactful performance tuning. We know that selecting the correct values for the OS, middleware layer, autoscaler, and cloud or container platform can cut costs, and massively improve performance. It just simply isn’t happening. Most teams select a large instance and leave the default for most OS and middleware settings. APM systems monitor basic app transactions and trigger alarms if something goes really wrong. Devs opt for the largest available VM to ensure peace of mind, even though the cost per instance will rack up over months.
Why? Though it would save them money, and drastically improve their performance, why aren’t companies optimizing their cloud apps?
This is why: It’s too damn complicated.
Even a simple application, comprising a few containers, will have trillions of resource and parameter permutations. Tweaking a setting for one part of the app can have a cascade of effects through the whole system. To get it right, you need perfect knowledge of the whole infrastructure and a deep familiarity with the application workload. And on top of this, every time there’s an update, the release of new code means you have to start over.
The solution? Let an AI tune your cloud apps for you.
Achieving maximum efficiency for cloud apps requires making judgments and decisions that are too complex and too overwhelming for any human staff member. At Opsani, we’ve leveraged an algorithmic technology called deep reinforcement learning to create a Continuous Optimization system. Opsani continuously examines millions of combinations of configurations, in order to identify the optimal combination of resources and parameter settings. It continually learns about the system, and its tweaks get better and better, as it uncovers counterintuitive solutions that would probably not occur to a human user.
Companies aren’t optimizing their cloud apps because they don’t know-how. The challenge is too complicated, the solution too mysterious.
But in the age of advanced AI, there’s no need to over-spend and over-provision. With Opsani, infrastructure can be tuned precisely to the workload and goals of the application. On average, when they implement Continuous Optimization, our customers experience a 2.5x increase in performance or a 40-70% decrease in cost.