Micro Benchmark Tests on "float" and "double" Operations
This chapter provides tutorial notes and example codes on micro benchmark tests on 'float' and 'double' data type operations using HotSpot JVM with and without JIT compilation. Topics include empty loop, assignment loop, add operation, multiplication operation and division operation.
These sections are omitted from this Web preview version. To view the full content,
see information on how to obtain the full version this book.
Benchmark Test Methods for "float" Operation
"float" Operations without JIT Compilation
"float" Operations with JIT Compilation
Benchmark Test Methods for "double" Operation
"double" Operations without JIT Compilation
"double" Operations with JIT Compilation
Performance of JIT on "float/double" Operations
"float/double" Operation Comparison between Systems
- HotSpot JIT compilation reduces execution time by about 10 times on some simple "float" or "double" operations.
- In JIT compilation mode, "double" operations are not much slower than "float" operations.
- Above results were collected on my Windows 7 computer with
Intel Core i5-6200U CPU @ 2.30Ghz with 4.00 GB of RAM.
Table of Contents
About This Book
JVM (Java Virtual Machine) Specification
Java HotSpot VM - JVM by Oracle/Sun
java.lang.Runtime Class - The JVM Instance
java.lang.System Class - The Operating System
ClassLoader Class - Class Loaders
Class Class - Class Reflections
JVM Runtime Data Areas
JVM Stack, Frame and Stack Overflow
Thread Testing Program and Result
CPU Impact of Multi-Thread Applications
I/O Impact of Multi-Thread Applications
CDS (Class Data Sharing)
Micro Benchmark Runner and JVM Options
Micro Benchmark Tests on "int" Operations
Micro Benchmark Tests on "long" Operations
Micro Benchmark Tests in JIT Compilation Mode
►Micro Benchmark Tests on "float" and "double" Operations
OpenJ9 by Eclipse Foundation
JRockit JVM 28.2.7 by Oracle Corporation
Full Version in PDF/EPUB