French Firm Brews Parallel Java Offering

By Michael Feldman

July 8, 2010

Yet another software toolmaker has offered up its solution to the parallel programming crisis. This week, French software vendor Ateji released a Java solution for multicore CPUs and multiprocessor server environments. Ateji PX for Java is aimed at developers who want to take advantage of current and future computer architectures by moving their legacy codes into the parallel realm.

Now Java is not exactly the first language that comes to mind when you think of parallel programming. Although the language encompasses the thread model and the notion of concurrency, it’s up to the Java Virtual Machine (JVM) component to implement true parallelism on the hardware. Attempts to add more parallel smarts to the language itself are numerous, but mostly incomplete. These include Titanium (a parallel dialect of Java), JOMP (an OpenMP-supported implementation), and JavaParty (for distributed computing environments), among others. Despite all these research projects, commercial adoption of parallel Java implementations never took hold.

Ateji, of course, would like to change that. One thing in its favor is that Ateji PX supports a wide range of parallel models, including task parallelism, data parallelism, speculative parallelism, recursive parallelism and distributed parallelism. And it does so with a relatively-small number of extensions to the standard language. The main addition is the “||” operator, which is used to tell the compiler to parallelize the associated source statement. For example, if you wanted to parallelize matrix multiplication via a for-loop you embed the || after the keyword:

for || (int i : I) {
   for (int j : J) {
      for (int k : K) {
         C[i][j] += A[i][k] * B[k][j];
      }
   }
}

Using this method, Ateji demonstrated a 3.4x speedup on a 4-core PC and a 12.5x speedup on a 16-core server. You could implement the equivalent parallelism by using Java’s thread library, but that would take about four times as much user code. Also, in that case, the original matrix multiplication algorithm gets hidden within a lot of thread management code.

In fact, the Ateji solution tosses the whole notion of threads out the window. “Threads are very ill-suited as a programming concept,” says the Ateji PX white paper. According to a Berkeley technical report cited in the whitepaper, a thread is really a hardware concept associated with the underlying processor architecture, rather than a natural language construct. This makes the thread model a much less productive way to think about parallelizing algorithms and can lead to buggy programs.

Instead, the Ateji model introduces the concept of a “branch,” which is basically equivalent to a parallelized statement. Beneath the covers, branches do get mapped to processor threads since the hardware demands it. But the mapping is context-specific, and the programmer can’t assume that the branch is (or is not) running in its own thread.

Ateji PX does provide compatibility with standard Java. The compiler front-end does a source-to-source translation to the base language, allowing the developer to keep his or her software development tool chain intact. It also enables the developer to revert back to vanilla Java once the parallelism has been implemented.

Even though the Ateji solution is being promoted as “parallel programming made simple,” the programmer is still stuck with the task of figuring out which statements and code sequences can be parallelized. The compiler doesn’t protect you from things like data races and deadlocks, so locks and atomic operations must be used if data is to be shared across parallelized code. For cases where sharing memory becomes too arduous or is just not available, Ateji PX also offers distributed memory parallelism. In fact, the Ateji PX programming manual suggests you mix the two, using a lot of small, well-defined parallel tasks that communicate via message passing.

For distributed parallelism, Ateji has come up with message passing primitives. At the simplest level, the programmer just needs to declare a channel object for message passing that is visible to both the sender and the receiver (using || chan ! value to send a message and || chan ? value to receive it). The fact that message passing can be specified at the language level rather than by invoking runtime routines (i.e., an MPI library) means that the source can be mapped to different types of distributed computing architecture independent of library implementations. It also allows different kinds of distributed models to be used — dataflow, stream programming, the Actor model, and the MapReduce algorithm.

Apparently, the distributed parallel feature is still a work in progress. According to the Ateji Web site, a version that is able to implement parallel branches on distributed memory hardware — compute clusters, supercomputers, grids and clouds — is still under development.

Of course, the whole idea of parallelizing code is to boost execution performance. Mapping code to more than one core or processor is bound to speed up the program, but Java, being an interpreted language, is not known for its stellar performance. For optimal execution, some heavy lifting will have to be done by the JVM. Just in time (JIT) compilers have helped to some extent, but the virtual machine model tends to be at odds with compute-bound codes. A 2004 report looked at the current state of Java for HPC and found that it is possible to deliver comparable performance to that of a Fortran/MPI implementation, for at least some codes, but scalability issues still remain. It’s not inconceivable that back-end Java technology could be developed for the performance market.

Although the official launch of Ateji PX was this week, the solution has been previewed by selected customers over the past year. A group at ISIMA claimed a seven-fold speedup with the Game of Life algorithm using Ateji PX on an 8-core Nehalem system. Another researcher at the University of Pisa benchmarked different parallel implementations of the convolution algorithm using the Ateji solution. In another case, an investment bank was able to parallelize a back-office Java app, reducing execution time from 40 minutes to eight minutes.

A free 30-day evaluation license is available for users that want to give Ateji PX a whirl. The solution is provided as an Eclipse plug-in, and all documentation and samples are provided online through the Ateji Web site.

Subscribe to HPCwire's Weekly Update!

Be the most informed person in the room! Stay ahead of the tech trends with industry updates delivered to you every week!

Q&A with Nvidia’s Chief of DGX Systems on the DGX-GB200 Rack-scale System

March 27, 2024

Pictures of Nvidia's new flagship mega-server, the DGX GB200, on the GTC show floor got favorable reactions on social media for the sheer amount of computing power it brings to artificial intelligence.  Nvidia's DGX Read more…

Call for Participation in Workshop on Potential NSF CISE Quantum Initiative

March 26, 2024

Editor’s Note: Next month there will be a workshop to discuss what a quantum initiative led by NSF’s Computer, Information Science and Engineering (CISE) directorate could entail. The details are posted below in a Ca Read more…

Waseda U. Researchers Reports New Quantum Algorithm for Speeding Optimization

March 25, 2024

Optimization problems cover a wide range of applications and are often cited as good candidates for quantum computing. However, the execution time for constrained combinatorial optimization applications on quantum device Read more…

NVLink: Faster Interconnects and Switches to Help Relieve Data Bottlenecks

March 25, 2024

Nvidia’s new Blackwell architecture may have stolen the show this week at the GPU Technology Conference in San Jose, California. But an emerging bottleneck at the network layer threatens to make bigger and brawnier pro Read more…

Who is David Blackwell?

March 22, 2024

During GTC24, co-founder and president of NVIDIA Jensen Huang unveiled the Blackwell GPU. This GPU itself is heavily optimized for AI work, boasting 192GB of HBM3E memory as well as the the ability to train 1 trillion pa Read more…

Nvidia Appoints Andy Grant as EMEA Director of Supercomputing, Higher Education, and AI

March 22, 2024

Nvidia recently appointed Andy Grant as Director, Supercomputing, Higher Education, and AI for Europe, the Middle East, and Africa (EMEA). With over 25 years of high-performance computing (HPC) experience, Grant brings a Read more…

Q&A with Nvidia’s Chief of DGX Systems on the DGX-GB200 Rack-scale System

March 27, 2024

Pictures of Nvidia's new flagship mega-server, the DGX GB200, on the GTC show floor got favorable reactions on social media for the sheer amount of computing po Read more…

NVLink: Faster Interconnects and Switches to Help Relieve Data Bottlenecks

March 25, 2024

Nvidia’s new Blackwell architecture may have stolen the show this week at the GPU Technology Conference in San Jose, California. But an emerging bottleneck at Read more…

Who is David Blackwell?

March 22, 2024

During GTC24, co-founder and president of NVIDIA Jensen Huang unveiled the Blackwell GPU. This GPU itself is heavily optimized for AI work, boasting 192GB of HB Read more…

Nvidia Looks to Accelerate GenAI Adoption with NIM

March 19, 2024

Today at the GPU Technology Conference, Nvidia launched a new offering aimed at helping customers quickly deploy their generative AI applications in a secure, s Read more…

The Generative AI Future Is Now, Nvidia’s Huang Says

March 19, 2024

We are in the early days of a transformative shift in how business gets done thanks to the advent of generative AI, according to Nvidia CEO and cofounder Jensen Read more…

Nvidia’s New Blackwell GPU Can Train AI Models with Trillions of Parameters

March 18, 2024

Nvidia's latest and fastest GPU, codenamed Blackwell, is here and will underpin the company's AI plans this year. The chip offers performance improvements from Read more…

Nvidia Showcases Quantum Cloud, Expanding Quantum Portfolio at GTC24

March 18, 2024

Nvidia’s barrage of quantum news at GTC24 this week includes new products, signature collaborations, and a new Nvidia Quantum Cloud for quantum developers. Wh Read more…

Houston We Have a Solution: Addressing the HPC and Tech Talent Gap

March 15, 2024

Generations of Houstonian teachers, counselors, and parents have either worked in the aerospace industry or know people who do - the prospect of entering the fi Read more…

Alibaba Shuts Down its Quantum Computing Effort

November 30, 2023

In case you missed it, China’s e-commerce giant Alibaba has shut down its quantum computing research effort. It’s not entirely clear what drove the change. Read more…

Nvidia H100: Are 550,000 GPUs Enough for This Year?

August 17, 2023

The GPU Squeeze continues to place a premium on Nvidia H100 GPUs. In a recent Financial Times article, Nvidia reports that it expects to ship 550,000 of its lat Read more…

Shutterstock 1285747942

AMD’s Horsepower-packed MI300X GPU Beats Nvidia’s Upcoming H200

December 7, 2023

AMD and Nvidia are locked in an AI performance battle – much like the gaming GPU performance clash the companies have waged for decades. AMD has claimed it Read more…

DoD Takes a Long View of Quantum Computing

December 19, 2023

Given the large sums tied to expensive weapon systems – think $100-million-plus per F-35 fighter – it’s easy to forget the U.S. Department of Defense is a Read more…

Synopsys Eats Ansys: Does HPC Get Indigestion?

February 8, 2024

Recently, it was announced that Synopsys is buying HPC tool developer Ansys. Started in Pittsburgh, Pa., in 1970 as Swanson Analysis Systems, Inc. (SASI) by John Swanson (and eventually renamed), Ansys serves the CAE (Computer Aided Engineering)/multiphysics engineering simulation market. Read more…

Choosing the Right GPU for LLM Inference and Training

December 11, 2023

Accelerating the training and inference processes of deep learning models is crucial for unleashing their true potential and NVIDIA GPUs have emerged as a game- Read more…

Intel’s Server and PC Chip Development Will Blur After 2025

January 15, 2024

Intel's dealing with much more than chip rivals breathing down its neck; it is simultaneously integrating a bevy of new technologies such as chiplets, artificia Read more…

Baidu Exits Quantum, Closely Following Alibaba’s Earlier Move

January 5, 2024

Reuters reported this week that Baidu, China’s giant e-commerce and services provider, is exiting the quantum computing development arena. Reuters reported � Read more…

Leading Solution Providers

Contributors

Comparing NVIDIA A100 and NVIDIA L40S: Which GPU is Ideal for AI and Graphics-Intensive Workloads?

October 30, 2023

With long lead times for the NVIDIA H100 and A100 GPUs, many organizations are looking at the new NVIDIA L40S GPU, which it’s a new GPU optimized for AI and g Read more…

Shutterstock 1179408610

Google Addresses the Mysteries of Its Hypercomputer 

December 28, 2023

When Google launched its Hypercomputer earlier this month (December 2023), the first reaction was, "Say what?" It turns out that the Hypercomputer is Google's t Read more…

AMD MI3000A

How AMD May Get Across the CUDA Moat

October 5, 2023

When discussing GenAI, the term "GPU" almost always enters the conversation and the topic often moves toward performance and access. Interestingly, the word "GPU" is assumed to mean "Nvidia" products. (As an aside, the popular Nvidia hardware used in GenAI are not technically... Read more…

Shutterstock 1606064203

Meta’s Zuckerberg Puts Its AI Future in the Hands of 600,000 GPUs

January 25, 2024

In under two minutes, Meta's CEO, Mark Zuckerberg, laid out the company's AI plans, which included a plan to build an artificial intelligence system with the eq Read more…

Google Introduces ‘Hypercomputer’ to Its AI Infrastructure

December 11, 2023

Google ran out of monikers to describe its new AI system released on December 7. Supercomputer perhaps wasn't an apt description, so it settled on Hypercomputer Read more…

China Is All In on a RISC-V Future

January 8, 2024

The state of RISC-V in China was discussed in a recent report released by the Jamestown Foundation, a Washington, D.C.-based think tank. The report, entitled "E Read more…

Intel Won’t Have a Xeon Max Chip with New Emerald Rapids CPU

December 14, 2023

As expected, Intel officially announced its 5th generation Xeon server chips codenamed Emerald Rapids at an event in New York City, where the focus was really o Read more…

IBM Quantum Summit: Two New QPUs, Upgraded Qiskit, 10-year Roadmap and More

December 4, 2023

IBM kicks off its annual Quantum Summit today and will announce a broad range of advances including its much-anticipated 1121-qubit Condor QPU, a smaller 133-qu Read more…

  • arrow
  • Click Here for More Headlines
  • arrow
HPCwire