Towards the measurement of tuple space performance

Fiedler, Daniel; Walcott, Kristen; Richardson, Tanya; Kapfhammer, Gregory M.; Amer, Ahmed; Chrysanthis, Panos K.

doi:10.1145/1111572.1111574

Cited by 10 publications

(6 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This will enable us to identify the impact that HotSpot TM adaptive optimization has on the performance of the communication mechanisms. Additional research will also investigate alternatives to socket and XML-RPC communication such as Java-MPI [11,19], Java RMI [12,25], tuple spaces [4,9,36], and JXTA [26,31]. These alternatives will be compared to sockets and XML-RPC in order to determine how each communication method compares to the others in terms of response time, network resource consumption, and JVM behavior.…”

Section: Discussionmentioning

confidence: 99%

An empirical comparison of Java remote communication primitives for intra-node data transmission

Burdette

Jones

Blose

et al. 2012

SIGMETRICS Perform. Eval. Rev.

Self Cite

View full text Add to dashboard Cite

This paper presents a benchmarking suite that measures the performance of using sockets and eXtensible Markup Language remote procedure calls (XML-RPC) to exchange intra-node messages between Java virtual machines (JVMs). The paper also reports on an empirical study comparing sockets and XML-RPC with response time measurements from timers that use both operating system tools and Java language instrumentation. By leveraging packet filters inside the GNU/Linux kernel, the benchmark suite also calculates network resource consumption. Moreover, the framework interprets the response time results in light of memory subsystem metrics characterizing the behavior of the JVM. The empirical findings indicate that sockets perform better when transmitting small to very large objects, while XML-RPC exhibits lower response time than sockets with extremely large bulk data transfers. The experiments reveal trade-offs in performance and thus represent the first step towards determining if Java remote communication primitives can support the efficient exchange of intra-node messages.

show abstract

Section: Discussionmentioning

confidence: 99%

An empirical comparison of Java remote communication primitives for intra-node data transmission

Burdette

Jones

Blose

et al. 2012

SIGMETRICS Perform. Eval. Rev.

Self Cite

View full text Add to dashboard Cite

show abstract

“…The software environment includes Windows XP professional as an operating system, Java JDK 1.4.2\_04 [8], Jini(TM) Technology Starter Kit v2.0.2 [9] and a free visual platform for Jini 2.0 that is calledblitz-javaspace. Inca X(TM) [12].…”

Section: System Performance Testmentioning

confidence: 99%

High Available Fault Tolerant Technique in Distributed Shared Memory

Refaat¹,

Badawi²

2016

IJCA

View full text Add to dashboard Cite

ABSTRACTdistributed systems, that are based on constructing a network of heterogeneous computers, suffer from the problem of failing components during the system run time. In case of failure, the distributed applications must be restarted from the scratch. The main goal of this research is to add the dynamic failure recovery technique to the JavaSpaces server. So, the client continues its jobs while failures occur in the system. Also, the new technique in JavaSpaces is evaluated by analyzing and testing.

show abstract

“…Although the performance and scalability of the tuple space is crucial for the usefulness of LINDA-based coordination languages, for a long time the real performance of tuple spaces (and thus their suitability for HPC) remained doubtful. In 2005, Fiedler et al [20] presented SETTLE, an "approach for measuring the throughput and response time of a tuple space when it handles concurrent local space interactions".…”

Section: A Coordination Languages Tuple Spaces and Key-mentioning

confidence: 99%

“…The SETTLE [20] framework is used to assess the performance of the Cwmwl tuple space. SETTLE assesses tuple space performance as execution time and throughput as a function of:  Number of (concurrent) clients.…”

Section: Performance Evaluationmentioning

confidence: 99%

Cwmwl, a LINDA-based PaaS Fabric for the Cloud

Fritsch

Walker²

2014

JCM

View full text Add to dashboard Cite

In this paper we introduce a new Platform-as-a-Service cloud environment that combines the LINDA coordination language, an in-memory key-value store, with functional programming to facilitate efficient execution of tenant plugins and applications. In the implementation a tuple space plays a central role in introducing deterministic services for basic parallel programming, including message passing, persistent infinite message pools and transactions. Redis, a keyvalue store, serves as the in-memory tuple space that glues together parallel constructs (i.e. skeletons) of formerly monolithic business applications to form an elastic distributed application. Although functional programming languages have adopted new runtime technology to achieve parallel execution , which is mostly focused on threads, it rarely offers an obvious way to match functions to threads. We find that the LINDA tuple space and its coordination model offers a general purpose paradigm to tackle synchronisation issues that ties into both domains of computing clouds: computation through supporting common skeletons and big data (analytics) through serving as an in-memory data grid.

show abstract

Towards the measurement of tuple space performance

Cited by 10 publications

References 19 publications

An empirical comparison of Java remote communication primitives for intra-node data transmission

An empirical comparison of Java remote communication primitives for intra-node data transmission

High Available Fault Tolerant Technique in Distributed Shared Memory

Cwmwl, a LINDA-based PaaS Fabric for the Cloud

Contact Info

Product

Resources

About