JProfiler Get rid of your performance problems and memory leaks! Modern Garbage Collection Tuning Java Performance Training Courses COURSES AVAILABLE NOW. We can provide training courses to handle all your Java performance needs Java Performance Tuning, 2nd ed The classic and most comprehensive book on tuning Java Java Performance Tuning Newsletter Your source of Java performance news. Subscribe now! Enter email: Training online Threading Essentials course JProfiler Get rid of your performance problems and memory leaks!

Back to newsletter 262 contents

https://medium.com/@hari.bodicherla/throttling-service-algorithms-explained-f5f14c13a752
Throttling service algorithms explained (Page last updated August 2022, Added 2022-09-28, Author hari krishna bodicherla, Publisher Medium). Tips:

• Most large tech companies enforce some form of API throttling to ensure service availability and to fight DDOS attacks.
• Common Algorithms for (per-user) throttling are: Leaking Bucket - a FIFO queue holds requests and whenever the queue is full new requests are dropped; Token Bucket - a fixed size bucket holds tokens filled at a fixed rate, requests need a token to proceed, requests unable to get a ticket are dropped; Fixed Window counter - allow N number of calls per time period; Sliding Window log - drop requests when the sliding window is full, holding timestamps in a log; Sliding window counter - CurrentRequestCount = RequestsCurrentWindow + RequestsPreviousWindow * PercentageOfOverlap, if the CurrentRequestCount is below the threshold, accept the request, otherwise drop it.

https://medium.com/javarevisited/five-api-performance-optimization-tricks-that-every-java-developer-must-know-75324ee1d244
Five API Performance Optimization Tricks that Every Java Developer Must Know (Page last updated August 2022, Added 2022-09-28, Author Lance_Ly, Publisher Javarevisited). Tips:

• Parallelize executions that are long running, especially calls to external services.
• Keep transactions as short as possible, especially avoid long running transactions as they lock out other requests.
• Ensure that large collections of data are indexed to access and update efficiently.
• Return data using paging where chunks of the data are returned one at a time as requested, so unread data is not provided.
• Cache for faster responses where that is needed.

https://www.youtube.com/watch?v=N5WbXprLoVM
Five most useful profiling techniques to use with modern Java apps (Page last updated July 2022, Added 2022-09-28, Author Laurentiu Spilca, Publisher Voxxed). Tips:

• Monitor SQL queries to identify inefficient communication between the app and the DBMS - visualvm has a JDBC profiler plugin, and p6spy is another technology that lets you do that easily.
• Profile the application to find the execution bottlenecks and optimize these away.
• Call graphs can be a useful tool to track activity through application execution and understand the overall time sent in the application flow of execution.
• Analyze your app's memory usage using a memory profiler (most profilers have one, eg visualvm) - look for memory leaks and high allocation rates.
• Use sampling profiling before bytecode profiling, as sampling has a much lower overhead and often gives you enough information to optimize - except for reactive apps.
• Bottlenecks that have execution time significantly larger than CPU time, are spending their time waiting.

Jack Shirazi Back to newsletter 262 contents