The CMS collector uses an evacuating collector for the young generation and a mostly concurrent mark sweep (non-compacting) algorithm for the tenured. CMS young generation GC is very similar to the parallel collector. The CMS garbage collector also requires coordination between the young and tenured GC. This is why CMS uses a parallel collector […]
The main goal of concurrent collectors is to reduce GC pause times. They were developed due to the growing demands for reduced application pause time. GC threads run in conjunction with application threads doing most of the GC work before pausing. Truly concurrent collectors are very hard to achieve as mutator threads keep modifying […]
JVM GC Implementation The JVM GC is a pluggable subsystem that enables the JVM to execute the same Java program with different GC implementations. Using a new GC algorithm does not require any change to the existing Java program. However, the chosen GC implementation may significantly impact the program’s performance. Java provides three types […]
JVM GC Evolution The JVM’s GC algorithms have evolved over the past three decades. The very first GC algorithm shipped with the JVM was the serial collector. Initially, Java was popular on the client side (Java Applets), and the serial collector was perfect for its GC needs. However, the serial collector worked well only on small […]
JVM GC Terminology The terminology used to describe garbage collection algorithms can be confusing. In this section, we will explain some commonly used terms and their meanings. Live set – The live set refers to the number of live objects in a heap. The size of the live set has a significant impact on the […]
Weak Generational Hypothesis The weak generational hypothesis has had a profound impact on the JVM’s heaps layout. Understanding the weak generational hypothesis is essential in order to understand various GC algorithms and approaches. The Weak Generational Hypothesis states that most objects die young. In other words, most objects created will be garbage collected very quickly. […]
GC Mark, Sweep, and Compact Basics The Mark and Sweep algorithm is the basis for garbage collection in Java. Although the actual algorithms used by the JVM are considerably more complex, the mark and sweep algorithm forms the basis of garbage collection in the JVM and must be understood. As you might have guessed, there […]
Java is a popular language for business and enterprise computing. Java and the JVM have enjoyed enormous success over the last two decades. Java was initially considered a slow language (late 90’s). That changed with the introduction of the HotSpot virtual machine in April 99. HotSpot improved performance via “just in time” compilation and adaptive optimisation. Since […]
The CAP theorem is a tool used to makes system designers aware of trade-offs while designing networked shared-data systems. CAP has influenced the design of many distributed data systems. It made designers aware of a wide range of tradeoff to consider while designing distributed data systems. Over the year the CAP theorem has been widely […]
Sometimes an Apache Cassandra cluster can end up in an unbalanced state. An unbalanced state is where data is unevenly distributed across a cluster or locally configured data directories. There are a number of reasons this can happen. In this blog post, I will cover two basics reasons this might happen. A cluster can end up […]