Understanding Java Object Creation, JVM Memory, and Class Loading

The article begins by showing the simplest way to create an object in Java using Dog dog = new Dog(); and notes that other techniques such as reflection, deserialization, or cloning also exist.

It then introduces the JVM memory model, distinguishing stack memory (stores primitive values and references) from heap memory (stores actual objects and arrays, managed by garbage collection). An illustration of the two areas is provided.

Next, the process of compiling a Java source file into a .class byte‑code file is described. The generated byte‑code is loaded into memory by a ClassLoader , which creates a Class object that describes the class.

The article explains the role of the Class object, showing how it holds metadata such as class name, methods, and fields, and how it can be used for reflection. A snippet of the Class source code, including the newInstance() method, is presented.

It compares Class.newInstance() (which can only invoke a no‑arg constructor and requires the class to be already loaded) with the new keyword (which directly creates an instance without those constraints). A table of differences is given, followed by examples: Class c = Class.forName("com.service.ClassAnalysis.Dog"); Dog dog = (Dog) c.newInstance(); // throws InstantiationException if no no‑arg constructor Constructor cs = Dog.class.getConstructor(String.class); Dog dog = (Dog) cs.newInstance("XiaoHei"); // works with parameterized constructor

The article then outlines the three phases of object creation after a class is loaded: (1) memory allocation for the object (including fields from super‑classes), (2) execution of initialization code (static blocks first, then instance blocks and constructors), and (3) linking the stack reference to the heap object.

Finally, it highlights three essential concepts for mastering Java internals: the class loader mechanism, the Class and Class object model, and the distinction between stack and heap memory for primitive versus reference types. Understanding these concepts aids later topics such as reflection, dynamic proxies, and framework source code analysis.