Fast Software Performance

Here are some aspects of software development that should be considered when trying to improve an application's performance.

Write Better Algorithms

Ultimately, the performance of an application is based on how well is is written. If the program loops around all elements in an array, the JVM will optimize the array bounds-checking so that the loop runs faster, and it may unroll the loop operations to provide an additional speedup. But if the purpose of the loop is to find a specific item, no optimization in the world is going to make the array-based code as fast as a different version that uses a HashMap.

A good algorithm is the most important thing when it comes to fast performance.

Improving your Algorithms & Data Structure Skills | medium

Write Less Code

A small well-written program will run faster than a large well-written program. This is true in general of all programs, and it applies specifically to Java programs. The more code that has to be compiled, the longer it will take until that code runs quickly. The more objects that have to be allocated and deallocated, the more work the garbage collector will have to do. If the objects are allocated and retained, the longer a garbage collection cycle will take. The more classes that have to be loaded from disk in the JVM, the longer it will take for a program to start. The more code that is executed, the less likely it will fit in the hardware caches on the computer. And the more code that has to be executed, the longer it will take.

More code / more features is a "Death by a thousand cuts". A developer will add a very small check for a feature then other developers will do the same and suddenly the performance has regressed. The cycle is repeated in the next release, and the performance starts to be noticeable. A couple of times during the process, performance testing might hit some resource threshold - a critical point in memory use, or a code cache overflow, etc. In those cases, the regular performance tests might make it possible to fix what appears to be a major regression. But over time, as the small regressions creep in, it will be harder and harder to fix.

I'm not advocating that you never add new features or code. The benefits of enhancing your applications is clear. But be aware of the small trade-offs you are making.

Writing Less Damn Code | heydonworks
Write Less Code | codeahoy
The Best Code Is No Code At All | codinghorror

Optimization Best Practices

Software engineers are expected to be familiar with the coding constructs that are known to be bad for performance. When writing code it is not "premature optimization" to apply your knowledge of best practices. So long as you are able to write clean, straightforward code that is simple to read and understand then you don't need to worry that you are prematurely optimizing your code.

30 Java Programming Tips and Best Practices for Beginners | javacodegeeks

Look Elsewhere

If you are developing standalone applications that use no external resources, then the performance of that application is mostly all that matters. Once an external resource such as a database is added, then the performance of both programs is important. And in a distributed environment, say an application server, a load balancer, a database, and more, then the performance of the single application server might be the least of the performance issues. In such an environment, a structured approach must be taken towards all aspects of the system. CPU usage, I/O latencies, and throughput of all parts of the system must be measured and analyzed; only then can it be determined which component is causing the performance bottleneck.

Optimize for the Common Case

It is tempting to treat all performance aspects as equally important. But focus should be given to the common use case scenarios.

• Optimize code by profiling it with realistic usage and focusing on the operations taking the most time.
• Apply Occam's razor to diagnosing performance problems. The simplest explanation for a performance issue is the most conceivable cause; a performance bug in new code is more likely than a bug in the JVM or the operating system.
• Write simple algorithms for the most common operations in the application. In general, optimize for the majority of your users. That will give you good overall performance stats but might not be what you need. For example, if 10% of your users, make 90% of your profit then you should optimize for that 10%.

Finally, JVM Tuning

Now that you have gone through the previous aspects that affect program performance, you are in a good place to tune the JVM.