Using Method R to write high-performance applications

Imagine the scene: you’ve spent the best part of a year developing a new, strategically important application for your business. It’s now launch day. Your directors and C-level officers are gathered in the room for the big reveal, perhaps alongside some potential customers. You fire up the application, ready to demonstrate what your team has built. You click. And then you wait. And wait some more. People start to fidget. The CIO frowns. You struggle through the demo, but the awkwardness of the situation gets worse, as everyone in the room realizes nobody in their right mind is going to tolerate software this slow.

A common challenge

We hear this story all too often. It happens when performance testing isn’t integrated into an application’s development cycle. Consequently, you only discover problems when it’s too late to do anything about them. People often leave ‘high-performance as a feature’ out of their applications because they only perceive performance as a consequence of other practices. This is why countless development teams around the world simply release their software and cross their fingers in hope that everything performs well. This creates two problems: the software has no performance-measurement capability built-in, and (mostly owning to the first deficiency) the software itself isn’t performance-optimized. You can’t afford to leave performance to chance: it has too big an impact on your reputation and profits. The good news is that you don’t have to, because performance is something you can measure and control.

Early performance testing

You need to know the performance characteristics of the software you’re building, from the day you begin building it. Using Method R Workbench, you can make measuring, managing and optimizing performance part of your development processes from the outset. Method R Workbench makes it easy for your development teams to test their code as they go along (and it’s especially easy if they’re using Oracle SQL Developer). This means they can spot performance problems with their code as they write it.

Catching problems before they cause trouble

Most application developers aren’t Oracle Database experts. Project managers may try to compensate by enforcing strict development standards, but no list of ‘best practices for Oracle developers’ is good enough. The way round this knowledge gap is to get feedback. Feedback is key to any improvement process. Shorter feedback loops mean greater agility. Rapid feedback from Method R Workbench stimulates developers of all experience levels to ask the right questions about their code as they produce it. This will enable them to understand the performance impact of every decision they make, thereby removing potential performance issues and ensuring their code is as efficient as possible (the key to scalability) before it leaves their machines.

Measuring and managing performance from the start has numerous benefits, not least that the earlier you catch a problem, the quicker, cheaper and less embarrassing it is to resolve.

Knowing your application’s limits

Before you release an application into the wider world, you need to know its performance limits. Is it sufficiently robust to perform well under the loads it’s likely to experience? Method R Workbench helps you find what these limits are, by enabling you to accurately measure your application under different conditions. The ability to see where time is going enables your developers to direct their attention to the aspects of the code that pose the greatest risk to scalability.

The process is straightforward. First, measure how the application spends its time. These measurements show where the code is most vulnerable. Knowing these vulnerabilities helps your developers create the right tests that will show your team exactly where the application will break down. If the only way you can make performance deteriorate to unacceptable levels is by pushing the application far beyond the limits it’ll face in production, you know you’re safe. However, if performance drops below acceptable levels before you reach your expected peak loads, you need to make changes to fix this. Your ability to measure your program’s time is what gives you the proper focus.

Enabling developers and DBAs to work together

Method R Workbench changes the culture of IT organizations by improving collaboration among users, developers, Oracle DBAs and system administrators, because it provides objective results that all parties can understand. It helps teams work together to tackle inefficiencies anywhere in a system, including UX and UI design, data and application design, application code, and system and database configuration. Importantly, it means your developers can write optimized code to work with Oracle Database, without needing to understand everything about Oracle all at once.

A transformational difference

Method R Workbench transforms how businesses create and release applications. It helps increase quality and reduce costs. It speeds up the overall lifecycle and reduces the number of performance problems that aren’t detected until late-stage testing or afterwards. Perhaps most importantly of all, it minimizes embarrassment in front of C-level executives or potential customers, protecting people’s jobs, and businesses’ reputations and profits.

Insights