Welcome!

IBM Cloud Authors: Yeshim Deniz, Elizabeth White, Pat Romanski, Liz McMillan, Stefan Bernbo

Related Topics: Java IoT, IBM Cloud, Weblogic

Java IoT: Article

Java vs C++ "Shootout" Revisited

Java vs C++ "Shootout" Revisited

Keith Lea writes of the benchmark, on his results page, "I was sick of hearing people say Java was slow, when I know it's pretty fast, so I took the benchmark code for C++ and Java from the now outdated Great Computer Language Shootout and ran the tests myself."

Lea used G++ (GCC) 3.3.1 20030930 (with glibc 2.3.2-98) for the C++, with the -O2 flag (for both i386 and i686). He compiled the Java code normally with the Sun Java 1.4.2_01 compiler, and ran it with the Sun 1.4.2_01 JVM. He ran the tests on Red Hat Linux 9 / Fedora Test1 with the 2.4.20-20.9 kernel on a T30 laptop. The laptop "has a Pentium 4 mobile chip, 512MB of memory, a sort of slow disk," he notes.

The results he got were that Java is significantly faster than optimized C++ in many cases.

"They also show that no one should ever run the client JVM when given the choice," Lea adds. ("Everyone has the choice," he says. To run the server VM, see instructions in the Using the Server JVM section below.)

JDJ has agreed to post online anyone else's results as long as they use Java 1.4.2 or higher and any version of GCC that produces faster or equivalent code than the 3.3.1 I used. We encourage you to download the source and/or the binaries and perform the tests yourself, with your favorite compiler and on your favorite platform.


Lea's Data and Results

JVM startup time was included in these results. "That means even with JVM startup time, Java is still faster than C++ in many of these tests," says Lea.

Some of the C++ tests would not compile. "I've never been very good at decoding GCC's error messages," he admits, "so if I couldn't fix a test with a trivial modification, I didn't include it in my benchmarks."

Lea also modified one of the tests, the string concatenation test for Java.

"The test was creating a new StringBuffer in each iteration of the loop, which was just silly," he explains. "I updated the code to use a single StringBuffer and appending to it inside the loop."

(The updated tests at the original shootout use this new method.)

"Java lost this benchmark even with the modifications," Lea declares. "So if anyone wants to accuse me of biasing the results, they're going to have to try harder."

Several versions of some of the C++ tests (like matrix) were present in the original shootout source, he continues. 

"I used the versions without numbers in them, like matrix.g++ instead of matrix.g++2. I don't know which of these were used in the original benchmarks, but from my quick experimenting, the numberless ones generally ran faster than their numbered counterparts."

"Looking at them again," Lea says, "matrix.g++3 runs faster than the matrix.g++ that I use. However, it still runs slower than the Java version, so I don't plan to modify the graph/data unless someone asks me to, since getting that graph in the first place was sort of a pain.)"

He continues: "I've been told that the C++ code for the Method Call benchmark returns by value while the Java code returns by reference, and that modifying the C++ code to pass a pointer makes that benchmark faster. However, even with the modification, the C++ version still runs slower than the Java version."

Lea ran th Java and the C++ tests to "warm up" (both the Java and C++ tests got faster after he ran them a few times).

"I've been told that these tests are invalid because they were run with GCC," he concedes, adding: "I have seen both benchmarks that show GCC producing faster code than Visual Studio's VC++ compiler, and benchmarks showing the opposite. If I update the benchmarks with another compiler added, it will be the Intel C++ Compiler, which I'm pretty sure produces faster code than VC++."

Lea says he's been accused of biasing the results by using the -O2 option for GCC, "supposedly because -O2 optimizes for space, thus slowing down the benchmark," he explains.

But this is not what -O2 does, he points out, referring to the GCC -O documentation:

JVM startup time was included in these results. "That means even with JVM startup time, Java is still faster than C++ in many of these tests," says Lea.

Some of the C++ tests would not compile. "I've never been very good at decoding GCC's error messages," he admits, "so if I couldn't fix a test with a trivial modification, I didn't include it in my benchmarks."

Lea also modified one of the tests, the string concatenation test for Java.

"The test was creating a new StringBuffer in each iteration of the loop, which was just silly," he explains. "I updated the code to use a single StringBuffer and appending to it inside the loop."

(The updated tests at the original shootout use this new method.)

"Java lost this benchmark even with the modifications," Lea declares. "So if anyone wants to accuse me of biasing the results, they're going to have to try harder."

Several versions of some of the C++ tests (like matrix) were present in the original shootout source, he continues. 

"I used the versions without numbers in them, like matrix.g++ instead of matrix.g++2. I don't know which of these were used in the original benchmarks, but from my quick experimenting, the numberless ones generally ran faster than their numbered counterparts."

"Looking at them again," Lea says, "matrix.g++3 runs faster than the matrix.g++ that I use. However, it still runs slower than the Java version, so I don't plan to modify the graph/data unless someone asks me to, since getting that graph in the first place was sort of a pain.)"

He continues: "I've been told that the C++ code for the Method Call benchmark returns by value while the Java code returns by reference, and that modifying the C++ code to pass a pointer makes that benchmark faster. However, even with the modification, the C++ version still runs slower than the Java version."

Lea ran the tests many times before running the "official" recorded set of tests, so there was plenty of time for both Java and the C++ tests to "warm up" (both the Java and C++ tests got faster after he ran them a few times).

"I've been told that these tests are invalid because they were run with GCC," he concedes, adding: "I have seen both benchmarks that show GCC producing faster code than Visual Studio's VC++ compiler, and benchmarks showing the opposite. If I update the benchmarks with another compiler added, it will be the Intel C++ Compiler, which I'm pretty sure produces faster code than VC++."

Lea says he's been accused of biasing the results by using the -O2 option for GCC, "supposedly because -O2 optimizes for space, thus slowing down the benchmark," he explains.

But this is not what -O2 does, he points out, referring to the GCC -O documentation:

-O2: Optimize even more. GCC performs nearly all supported optimizations that do not involve a space-speed tradeoff. The compiler does not perform loop unrolling or function inlining when you specify -O2. As compared to -O, this option increases both compilation time and the performance of the generated code.

"On the other hand, -O3 performs space-speed tradeoffs, and -O performs fewer optimizations. Thus, for these tests, I think O2 was the best choice," Lea concludes.

 

"I don't have an automated means of building and benchmarking these things (and the scripts that came with the original shootout didn't run for me)," he continues. "I really do want people to test it on their own machines, but it's going to take some work, I guess."

Lea compiled the C++ code with:

g++ [test].cpp -O2 -march=i386 -o [test]-386

g++ [test].cpp -O2 -march=i686 -o [test]-686

and the Java code with:

javac [test].java

To see how he ran the binaries, see the run log. You can download the source code he used in either .bz2 or .zip format.

Using the Server JVM

Every form of Sun's Java runtime comes with both the "client VM" and the "server VM."

"Unfortunately, Java applications and applets run by default in the client VM," Lea observes. "The Server VM is much faster than the Client VM, but it has the downside of taking around 10% longer to start up, and it uses more memory."

Lea explains the two ways to run Java applications with the server VM as follows

  1. When launching a Java application from the command line, use java -server [arguments...] instead of java [arguments...]. For example, use java -server -jar beanshell.jar.
  2. Modify the jvm.cfg file in your Java installation. (It's a text file, so you can use Notepad or Emacs to edit it.) This is located in C:\Program Files\Java\j2reXXX\lib\i386\ on Windows, /usr/java/j2reXXX/lib/i386/ on Linux. You will see two lines:
    -client KNOWN
    -server KNOWN
    You should change them to:
    -server KNOWN
    -client KNOWN
    This change will cause the server VM to be run for all applications, unless they are run with the -client argument.

He can be contacted at

Every form of Sun's Java runtime comes with both the "client VM" and the "server VM."

"Unfortunately, Java applications and applets run by default in the client VM," Lea observes. "The Server VM is much faster than the Client VM, but it has the downside of taking around 10% longer to start up, and it uses more memory."

Lea explains the two ways to run Java applications with the server VM as follows

  1. When launching a Java application from the command line, use java -server [arguments...] instead of java [arguments...]. For example, use java -server -jar beanshell.jar.
  2. Modify the jvm.cfg file in your Java installation. (It's a text file, so you can use Notepad or Emacs to edit it.) This is located in C:\Program Files\Java\j2reXXX\lib\i386\ on Windows, /usr/java/j2reXXX/lib/i386/ on Linux. You will see two lines:
    -client KNOWN
    -server KNOWN
    You should change them to:
    -server KNOWN
    -client KNOWN
    This change will cause the server VM to be run for all applications, unless they are run with the -client argument.

He can be contacted at keith@kano.net.

Links

More Stories By Jeremy Geelan

Jeremy Geelan is Chairman & CEO of the 21st Century Internet Group, Inc. and an Executive Academy Member of the International Academy of Digital Arts & Sciences. Formerly he was President & COO at Cloud Expo, Inc. and Conference Chair of the worldwide Cloud Expo series. He appears regularly at conferences and trade shows, speaking to technology audiences across six continents. You can follow him on twitter: @jg21.

Comments (152) View Comments

Share your thoughts on this story.

Add your comment
You must be signed in to add a comment. Sign-in | Register

In accordance with our Comment Policy, we encourage comments that are on topic, relevant and to-the-point. We will remove comments that include profanity, personal attacks, racial slurs, threats of violence, or other inappropriate material that violates our Terms and Conditions, and will block users who make repeated violations. We ask all readers to expect diversity of opinion and to treat one another with dignity and respect.


Most Recent Comments
B Bennett 06/15/04 06:25:37 PM EDT

Mr Salarian and people with similar questions should consult the methodology notes from the original shootout (whose code was borrowed for this test) at

http://www.bagley.org/~doug/shootout/method.shtml

That should help clarify what exactly is getting tested, and by what methods.

A. Salarian 06/15/04 06:01:52 PM EDT

Sir, with all due respect I have to say your benchmark is worthless. You have note compared a single C++ program to a Java program. The codes you have tested do not show anything. Why not some real world situation on WHY people use C++ but not Java? Like some complex matrix operations using template based libraries like Blitz++ and equivalent Java industry level Matrix libs? Some applications with a large number of dynamic allocations and deallocations like a dynamic patricia tri in a LZW implementaion? Why not largescale FFTs to test the efficiency of cache management? Why you didn''t use the instrumented C++ (i.e. using the profiling guided compilations)?
You can get any result you wish as long as you don''t test any real world code...

IoT & Smart Cities Stories
The deluge of IoT sensor data collected from connected devices and the powerful AI required to make that data actionable are giving rise to a hybrid ecosystem in which cloud, on-prem and edge processes become interweaved. Attendees will learn how emerging composable infrastructure solutions deliver the adaptive architecture needed to manage this new data reality. Machine learning algorithms can better anticipate data storms and automate resources to support surges, including fully scalable GPU-c...
Machine learning has taken residence at our cities' cores and now we can finally have "smart cities." Cities are a collection of buildings made to provide the structure and safety necessary for people to function, create and survive. Buildings are a pool of ever-changing performance data from large automated systems such as heating and cooling to the people that live and work within them. Through machine learning, buildings can optimize performance, reduce costs, and improve occupant comfort by ...
The explosion of new web/cloud/IoT-based applications and the data they generate are transforming our world right before our eyes. In this rush to adopt these new technologies, organizations are often ignoring fundamental questions concerning who owns the data and failing to ask for permission to conduct invasive surveillance of their customers. Organizations that are not transparent about how their systems gather data telemetry without offering shared data ownership risk product rejection, regu...
René Bostic is the Technical VP of the IBM Cloud Unit in North America. Enjoying her career with IBM during the modern millennial technological era, she is an expert in cloud computing, DevOps and emerging cloud technologies such as Blockchain. Her strengths and core competencies include a proven record of accomplishments in consensus building at all levels to assess, plan, and implement enterprise and cloud computing solutions. René is a member of the Society of Women Engineers (SWE) and a m...
Poor data quality and analytics drive down business value. In fact, Gartner estimated that the average financial impact of poor data quality on organizations is $9.7 million per year. But bad data is much more than a cost center. By eroding trust in information, analytics and the business decisions based on these, it is a serious impediment to digital transformation.
Digital Transformation: Preparing Cloud & IoT Security for the Age of Artificial Intelligence. As automation and artificial intelligence (AI) power solution development and delivery, many businesses need to build backend cloud capabilities. Well-poised organizations, marketing smart devices with AI and BlockChain capabilities prepare to refine compliance and regulatory capabilities in 2018. Volumes of health, financial, technical and privacy data, along with tightening compliance requirements by...
Predicting the future has never been more challenging - not because of the lack of data but because of the flood of ungoverned and risk laden information. Microsoft states that 2.5 exabytes of data are created every day. Expectations and reliance on data are being pushed to the limits, as demands around hybrid options continue to grow.
Digital Transformation and Disruption, Amazon Style - What You Can Learn. Chris Kocher is a co-founder of Grey Heron, a management and strategic marketing consulting firm. He has 25+ years in both strategic and hands-on operating experience helping executives and investors build revenues and shareholder value. He has consulted with over 130 companies on innovating with new business models, product strategies and monetization. Chris has held management positions at HP and Symantec in addition to ...
Enterprises have taken advantage of IoT to achieve important revenue and cost advantages. What is less apparent is how incumbent enterprises operating at scale have, following success with IoT, built analytic, operations management and software development capabilities - ranging from autonomous vehicles to manageable robotics installations. They have embraced these capabilities as if they were Silicon Valley startups.
As IoT continues to increase momentum, so does the associated risk. Secure Device Lifecycle Management (DLM) is ranked as one of the most important technology areas of IoT. Driving this trend is the realization that secure support for IoT devices provides companies the ability to deliver high-quality, reliable, secure offerings faster, create new revenue streams, and reduce support costs, all while building a competitive advantage in their markets. In this session, we will use customer use cases...