TLAB mechanism of JVM object memory allocation and TLAB process in G1

One ： Review the life cycle of an object

1.1 Object lifecycle flowchart

1.2 Focus on

Object's finalize() Methods are called only once , That is to say, an object can only have one resurrection opportunity , But the resurrection of an object will destroy the life cycle of the object , Interfere with the JVM Collection of objects , So in finalize() Methods are generally used to release resources before collection of write objects .

---

Two : TLAB( Memory allocation buffer )

From the life cycle of the object in the above figure , We have learned that the memory allocation and initialization of an object must be performed before it is used , Next, we describe two strategies for object memory allocation

2.1 Why would there be TLAB

We know that object memory allocation was initially considered in the new generation Eden Area allocates memory for objects , But a heap is an area shared by threads , In a multithreaded environment, it is common for multiple threads to manipulate the same block of memory at the same time , To ensure the security of data during object allocation , We need to use global locks to solve the problem of multiple threads accessing the same block of memory at the same time , But global locking reduces efficiency and program throughput , To solve this problem TLAB emerge as the times require .

2.2 TLAB

TLAB Yes Eden The memory of the area is divided , Carve out a small area , This area is called the object memory allocation buffer , It allocates a space for each thread .

advantage :

1. Use TLAB It can avoid a series of thread safety problems .
2. It improves the efficiency and throughput of the program .

This kind of use TLAB The strategy for memory allocation is called Fast allocation strategy .

---

3、 ... and ：G1 In the source TLAB Strategy

3.1 brief introduction

When allocating space , In order to improve the JVM Operating efficiency , The critical area should be minimized , Avoid global locks .G1 In the common application scenarios of , There will be a large number of accessors executing at the same time , To reduce lock conflicts ,JVM Introduced TLAB（ Thread local allocation buffer Thread Local Allocation Buffer） Mechanism .

3.2 flow chart

According to the source code, the following flow chart can be obtained :

---

If it helps you , Don't forget to like your collection !