Java in the High Performance Computing arena: Research, practice and experience

Taboada, Guillermo L.; Ramos, Sabela; Expósito, Roberto R.; Touriño, Juan; Doallo, Ramón

doi:10.1016/j.scico.2011.06.002

Cited by 80 publications

(47 citation statements)

References 40 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The current trend of multicore computing emphasises the need for parallelism, multithreading and communication in general (Taboada et al, 2013). It is important to utilise these concepts to make software scalable (Slinn, 2012), and to make full use of multicore processors.…”

Section: Discussionmentioning

confidence: 99%

“…This means that communication can take place between multiple processes on a single machine. This takes the form of socket programming using network protocols (such as TCP and UDP), through the use of Java's standard socket API (Taboada, Ramos, Expósito, Touriño, & Doallo, 2013). Messages can be encapsulated into network packets and passed through the IP stack as necessary.…”

Section: Interprocess Communicationmentioning

confidence: 99%

“…Messages are passed to programs running in separate JVMs using the loopback network, thereby providing a form of IPC. Taboada et al (2013) state that Java's network support over TCP/UDP is an inefficient communication mechanism. In addition, research conducted by Wells (2010) also found it to be inefficient.…”

Section: Interprocess Communicationmentioning

confidence: 99%

“…Yet another form of distributed communication is the Java Message Service (JMS) which provides communication between applications (Oracle, 2016). These high-level distributed mechanisms provide a solution to IPC but, according to Taboada et al (2013), at the cost of poor performance.…”

Section: Interprocess Communicationmentioning

confidence: 99%

See 3 more Smart Citations

Interprocess Communication with Java in a Microsoft Windows Environment

Smith¹,

Wells²

2017

SACJ

View full text Add to dashboard Cite

The Java programming language provides a comprehensive set of multithreading programming techniques but currently lacks interprocess communication (IPC) facilities, other than slow socket-based communication mechanisms (which are intended primarily for distributed systems, not interprocess communication on a multicore or multiprocessor system). This is problematic due to the ubiquity of modern multicore processors, and the widespread use of Java as a programming language throughout the software development industry. This work aimed to address this problem by utilising Microsoft Windows’ native IPC mechanisms through a framework known as the Java Native Interface. This enabled the use of native C code that invoked the IPC mechanisms provided by Windows, which allowed successful synchronous communication between separate Java processes. The results obtained illustrate the performance dichotomy between socket-based communication and native IPC facilities, with Windows’ facilities providing significantly faster communication. Ultimately, these results show that there are far more effective communication structures available. In addition, this work presents generic considerations that may aid in the eventual design of a generic, platform-independent IPC system for the Java programming language. The fundamental considerations include shared memory with semaphore synchronisation, named pipes and a socket communication model.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Interprocess Communicationmentioning

confidence: 99%

Section: Interprocess Communicationmentioning

confidence: 99%

Section: Interprocess Communicationmentioning

confidence: 99%

See 2 more Smart Citations

Interprocess Communication with Java in a Microsoft Windows Environment

Smith¹,

Wells²

2017

SACJ

View full text Add to dashboard Cite

show abstract

“…33 C#, 1-in-14.29 C, and 1-in-16.67 C++ [21]. With all of these benefits and large pool of developers, it is no wonder Java is transitioning into a computation language as well [13], [14], [22], [23], sequentially allocated objects are not guaranteed to be in adjoining memory locations). Popular computational languages have the ability to reserve contiguous blocks of memory for primitive types, providing both speed and size benefits.…”

Section: Introductionmentioning

confidence: 99%

UPC: Large-Scale Memory Efficient Java Primitive Collections

Hylock¹

2016

JSW

View full text Add to dashboard Cite

Abstract:The Java programming language is versatile and robust, however, for large-scale applications with memory intensive processes, much is still to be desired. The public Java Development Kit (JDK) supports neither forcible object destruction nor large collection sizes (limited to 231-1) as found in choice computational languages such as Fortran and C/C++. Within Java's Hotspot, however, the non-public sun.misc.Unsafe class allows such features through basic off-heap functionality. The submitted UPC collections are fully integrated with Unsafe to provide large (i.e., 263-1), destructible arrays, lists, hash sets, hash maps, and matrices, consistent with advanced computational needs. Additionally, a customized version of OpenJDK 1.9 with bulk-operation Unsafe support and companion collections are provided. These tools are compared to Java and extent third-party primitive collections. Testing indicates UPC performs in a consistent to superior manner compared to the state-of-the-art, with greater improvements realized by way of the modified virtual machine.

show abstract

Nonblocking collectives for scalable Java communications

Ramos

Taboada

Expósito

et al. 2014

Concurrency and Computation

Self Cite

View full text Add to dashboard Cite

SummaryThis paper presents a Java implementation of the recently published MPI 3.0 nonblocking message passing collectives in order to analyze and assess the feasibility of taking advantage of these operations in shared memory systems using Java. Nonblocking collectives aim to exploit the overlapping between computation and communication for collective operations to increase scalability of message passing codes, as it has been carried out for nonblocking point‐to‐point primitives. This scalability has become crucial not only for clusters but also for shared memory systems because of the current trend of increasing the number of cores per chip, which is leading to the generalization of multi‐core and many‐core processors. Message passing libraries based on remote direct memory access, thread‐based progression, or implementing pure multi‐threading shared memory support could potentially benefit from the lack of imposed synchronization by nonblocking collectives. But, although the distributed memory scenario has been well studied, the shared memory one has not been tackled yet. Hence, nonblocking collectives support has been included in FastMPJ, a Message Passing in Java (MPJ) implementation, and evaluated on a representative shared memory system, obtaining significant improvements because of overlapping and lack of implicit synchronization, and with barely any overhead imposed over common blocking operations. Copyright © 2014 John Wiley & Sons, Ltd.

show abstract

Java in the High Performance Computing arena: Research, practice and experience

Cited by 80 publications

References 40 publications

Interprocess Communication with Java in a Microsoft Windows Environment

Interprocess Communication with Java in a Microsoft Windows Environment

UPC: Large-Scale Memory Efficient Java Primitive Collections

Nonblocking collectives for scalable Java communications

Contact Info

Product

Resources

About