Isolating the performance impacts of network interface cards through microbenchmarks

Always Best Packet Switching (ABPS) is a novel approach for wireless communications that enables mobile nodes, equipped with multiple network interface cards (NICs), to dynamically determine the most appropriate NIC to use. Using ABPS, a mobile node can seamlessly switch to a different NIC in order to get better performance, without causing communication interruptions at the application level. To make this possible, NICs are kept always active and a software monitor constantly probes the channels for available access points. While this ensures maximum connection availability, considerable energy may be wasted when no access points are available for a given NIC. In this paper we address this issue by investigating the use of an "oracle" able to provide information on network availability. This allows to dynamically switch on/off NICs based on reported availability, thus reducing the power consumption. We present a Markov model which allows us to estimate the impact of the oracle on the ABPS mechanism: results show that significant reduction in energy consumption can be achieved with minimal impact on connection availability. We conclude by describing a prototype implementation of the oracle based on Web services and geolocalization.

Performance of 1 and 10 Gigabit Ethernet cards with server quality motherboards

Increasing attention has been recently devoted to software routers based on off-the-shelf hardware and open-source operating systems running on personal computer (PC) architectures. Today's high-end PCs PCI shared buses fit into the multi-gigabit-per-second routing segment, for a price much lower than that of commercial routers. However, commercially available Network Interface Cards (NICs) lack programmability, and require not only packets to cross the PCI bus twice, but also to process them in software by the operating system (OS), reducing routing performance. In this paper we discuss the design of an FPGA-based NIC that permits to overcome the limitations of commercial NICs and provide a detailed description of its implementation.

Two critical issues impact the overall performance of Linux clusters based on Intel servers: inter-process communication latency and data throughput. Underlying both of these performance challenges is the inefficient use of computational power and server CPU cycles to process the network protocols. Today’s modern high-end Network Interface Cards (NICs) are equipped with an onboard CPU. In most cases, these CPU’s are only used by the vendor and are operated by a proprietary OS, which makes them inaccessible to the HPC application developer. In this paper we present a design and implementation of a framework for building high-performance networking applications. The framework consists of an embedded NIC Operating System with a specialized scheduler. The main challenge in developing such a scheduler is the lack of a preemption mechanism in most high-end NICs. Our scheduler provides finer-grained schedules than the alternatives. We have implemented several network applications, and were able to increase their throughput while decreasing the host’s CPU utilization.KeywordsSchedule AlgorithmHigh Performance ComputingPeriodic TaskNetwork Interface CardHigh Performance Computing ClusterThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Matching web services and client requirements in the form of goals is a significant challenge in the discovery of semantic web services. The most common but unsatisfactory approach to matching is set‐based, where both the client and web services declare what objects they require and what objects they can provide. Matching then becomes the simple task of comparing sets of objects. This approach is inadequate because it says nothing about the functionality required by the client or the functionality provided by the web service. As an alternative, we use the Frame Logic as implemented in Flora‐2 to specify web service capabilities and client requirements, including their preconditions, postconditions, and ontologies, implement a logic‐based discovery agent using Flora‐2, demonstrate its usefulness in a medical appointment making scenario, and show its efficiency both theoretically and by benchmarking. The result is an expressive yet concise representation scheme for semantic web services, and a practical, efficient, powerful, and fully implemented matching engine based purely on logical inference for web service discovery, with direct applicability to Web Service Modeling Ontology and Web Service Modeling Language, because both are based on Frame Logic.

Proposes a scheme to realize a high-performance communication facility using a commodity network. This scheme does not require any special hardware or hardware-specific device drivers in order to adapt to many kinds of network interface cards (NICs). In this scheme, a reliable lightweight network protocol is handled directly on a data link layer called by a network device driver. An interrupt reaping technique is proposed to eliminate the hardware interrupt overhead when an application waits for a message. PM/Ethernet, an instance of the scheme, is implemented on Linux with minimal modification to the Linux kernel, and existing network device drivers are used without any modification. Using Pentium III 500-MHz PCs on Packet Engine's G-NIC II Gigabit Ethernet NIC, it achieves 77.5 MB/s bandwidth and 37.6 /spl mu/s round-trip time latency compared to that of TCP/IP, which achieves 46.7 MB/s bandwidth and 89.6 /spl mu/s round-trip time latency. The NAS parallel benchmark IS results show that MPI on PM/Ethernet achieves 75% better performance than MPI on TCP/IP and is 7.8% slower than that of MPI on Myrinet PM.

In the Grid environment, user can manage and make use of web services, such as uploading or downloading web services. But when user wants to download one web services to run it in a local machine, one problem may occur: The web service can only be run on a special operating system, which is not the operating system that the user operates. To solve this problem, we need construct one middleware to emulate the running environment of the web service in the operating system that user operates. Middleware is the component created by the Operation System according to its metadata, and it is also the center from which the Internet technology extended. Component-based middleware has emerged as an important architectural component in modern distributed systems, such as XPCOM, UNO and the CORBA Component Model. Java Virtual Machine (JVM) is a middleware, aimed at making application cross platform to run. The goal of our research is to extend this work in order to exploit one component-based middleware platform that can make web services to run on different operating system platforms, such as Elastos, Windows and Linux operating system. In this paper, a new and efficient component model based on COM (ezCOM) is proposed. More specifically, this paper describes in detail how ezCOM can be used to construct a full component-based crossing Operating system middleware platform and also puts forward one new measure model to compare the performance of ezCOM, Wine-COM, XPCOM and UNO.

Web services are the best way to solve cross-platform, cross-language, loosely coupled distributed systems in web environment at present. But because of some non-functional attributes such as authorization, security, transaction, logging, etc. in web services, the adaptation and reuse of web services are decreased. It is difficult to maintain and extend SOA-based applications which are composed of them. In order to solve the above problems, Aspect-Oriented Programming is introduced to web services in this paper, the non-functional attributes are separated and realized as aspectual services and the web services and aspectual services are weaved at SOAP level to construct the final system when the application runs.

While both SOAP and REST have been used widely to implement Web services and software integration, over time REST has emerged as the predominant approach. REST provides developers with a lower barrier to entry for implementation and greater development flexibility than SOAP. Its architectural conventions and best practices can be integrated into Web services incrementally as opposed to the all-or-nothing adoption of SOAP. In order to achieve generality, SOAP standards are extensive, rigid, and complex. This complexity can lead to implementations that introduce significant overhead on the network bandwidth consumption, execution times, and throughput of SOAP services, especially in the emerging resource-restricted mobile realm. This chapter provides an overview of the logical and physical design of modern Web services and discusses the strengths and weaknesses of the predominant styles. It provides evidence and reasoning behind the emergence of REST as the leader for the development of next-generation Web APIs and services. The chapter also delineates the key technologies that underlie REST and describes emerging and future research directions in support of REST-based APIs and service development.

One of the major challenges of the Service Oriented Computing (SoC) paradigm is the Web service discovery. Yet the mechanisms used to match Web services can be improved. However, considering the semantics of the Web service descriptions is a must for automating the process of discovering Web services. We have proposed a semantic-based Web service discovery algorithm that consists of two stages: The registry filtering stage and the descriptions matching stage. The aim of the first stage is to pick the advertisements that are relevant to the client request and ignore, from an early stage and before checking the details of the descriptions, the advertisements that are not able to satisfy the client request. However, the aim of the second stage is to check whether or not the relevant advertisements resulted from the registry filtering stage are actually able to satisfy the client query. This paper concerns the first stage. More precisely, we propose a semantic-based Web service registry filtering mechanism that takes the responsibility of narrowing down the number of Web service descriptions to be checked in detail to the number of only the relevant advertisements to the client request. We also propose an extension to Web Ontology Language for Services (OWL-S). With this extension, filtering the Web service registry based on the semantics of the available descriptions is possible.KeywordsClient RequestGlobal CategoryClient QueryRelevant AdvertisementThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Mobile technology applications have proven more interesting and more capable with each passing year, and continue to be a technology with new surprises. In this paper, we present a software development for weather report which is motivated as a cell phone programme application for web service. Users can get the latest weather information without accessing to the specific weather website. In addition, weather information can be deployed on any cell phone with internet explorer. Web Services, MIDlet programming is the technology applied in this paper. Weather report is designed as a two-tier structure, web application server and client. Web services are generated and published on the web. On client side, user, on the MIDP platform, can interact with web services with the agreement on SOAP protocol. In this paper, how to deploy web services, and the differences and constraints of MIDlet programming are presented.

As the amount of pathway information for various organisms is increasing very rapidly, performing various analyses on the full network of pathways for even multiple organisms can be possible and therefore developing an integrated database for storing and analyzing pathway information is becoming a critical issue. Until now analyzing these networks is not easy because of the nature of the existing pathway databases, which are often heterogeneous, incomplete, and/or inconsistent. We presented a database system called SPDBS to solve this problem. However, application-oriented systems like SPDBS have some limitations on the extension and integration of the heterogeneous databases. In this paper, we extend previous SPDBS into a web service prototype (SPDBSW) where all functions can be serviced on the web environment. The web services include pathway database integration/search, import/export of SBML documents, pathway reconstruction/visualization. SPDBSW has been implemented by the combination SPDBS and external web services such as OLS, KEGG and NCBI. And user can get more confidential and delicate information from KEGG or NCBI through their web services. The system can be extended or modified immediately by replacing its component web services. We provide SPDBSW at the website http://database.chungbuk.ac.kr/SPDBSW.

In this paper, we present a service-oriented architecture (SOA) based system to meet the on-demand VMI requirements. SOA is used to create flexible and efficient loosely-coupled systems . We utilize technologies including web services, SOAP, HTTP, XML and WSDL to develop a VMI system. Web services can support the integration of information and services across platforms and operating systems. In addition, we implement web service coordination (WS-Coordination) to enable coordinated transactions across platforms. WS-Coordination is a specification for providing protocols that coordinate the actions of distributed applications. Also, we present workflow of VMI operations from two aspects: seller¿s market and buyer¿s market. Finally, we use the business process execution language (BPEL) for web services to describe business processes of VMI. Service components are deployed flexibly in web environment using the Apache Axis2 web services engine.

A Web-Based Perioperative Dashboard as a Platform for Anesthesia Informatics Innovation.

We design a specific Ethernet network interface card (NIC) for accelerating video delivery by offloading the overheads of protocol headers identification/appending and CRC/checksums calculation, and speeding video bit streams with a dedicated video interface. Compared with the same operations of a 50 MHz ARM micro-controller, the NIC system saves 47,000 ns per frame. This NIC card also supports the coexistence of the IPv4 and IPv6 standard for the future extension. Both FPGA prototyping and 0.35 /spl mu/m cell-based design of the specific NIC system are given.