Milorad Tosic

A heuristic for scheduling task graphs with communication delays onto multiprocessors

Abstract The multiprocessor scheduling problem can be stated as finding a schedule for a task graph to be executed on a multiprocessor architecture so that the execution time can be minimized. Since this problem is known to be NP-hard, in all but a few very restricted cases, the main research efforts in this area are focused on heuristic methods for obtaining near-optimal solutions in a reasonable amount of time. A new compile-time single-pass multiprocessor scheduling technique, called chaining , has been developed and is presented in this paper. Chaining takes into account the communication overhead and can be applied to scheduling task graphs onto fully-connected multiprocessor architectures containing an arbitrary (bounded as well as an unbounded) number of processors. This technique can be viewed as a generalized list scheduling concept, that does not impose any preconditions about the ordering in which tasks are selected for scheduling as well as about the position within the current partial schedule where selected task can be placed. Varying the selection policy, implemented in this technique, we are able to generate a class of scheduling algorithms. As a representative example of this class we present Task Selection First (TSF) scheduler. We compare performances of the TSF scheduler with the dynamic level scheduler proposed by Sih and Lee, the dominant sequence clustering algorithm proposed by Yang and Gerasoulis, and the DSC/MLS algorithm, a modified version of Sarkars two-step scheduling technique.

A Compile-Time Scheduling Heuristic for Multiprocessor Architectures

The multiprocessor scheduling problem can be stated as finding a schedule for a task graph to be executed on a multiprocessor architecture so that the execution time can be minimized. Since this problem is known to be NP-hard, in all but a few very restricted cases, the main research efforts in this area are focused on heuristic methods for obtaining near-optimal solutions in a reasonable amount of time. A new compile-time single-pass multiprocessor scneduling technique, called chaining, has been developed and is presented in this paper. Chaining can be used to schedule task graphs onto multiprocessor architectures that contain an arbitrary number of processors connected in an irregular fashion, taking into account the expected execution and communication requirements of the task graph on the given multiprocessor architecture. This technique can be viewed as a generalization of the list scheduling technique, that does not impose any preconditions about the ordering according to which tasks are selected for scheduling. Varying the selection criteria, implemented in this technique, we have generated a new class of scheduling algorithms. An evaluation of this class was made on 360 randomly generated examples, and the estimated performances were compared with two list scheduling algorithms, the dynamic level scheduler proposed by Sih and Lee, and the earliest task first algorithm proposed by Hwang et al.

Trust-based peer assessment for virtual learning systems

In order to deal with new requirements imposed by emerging learning environments following social computing paradigm, we address the problem of assessment of individual student skills, contributions, and activities. Rather then clicking links to launch tools or to view content, such learning environments encourage more of a monitoring mode of operation that is very difficult to sense and record by the software alone. In this paper we propose adoption of peer-assessment approach in order to overcome the obstacle as well as to make the overall solution scalable. We propose a novel method for students peer-assessment based on trust concept. The overall approach is presented and practical experiments are conducted using developed web service. The grade scores determined by the learning peers/students are statistically proven as highly correlated with those marked by the teachers, indicating that approach proposed in this paper may be adopted as a legitimate assessment method.

Resource specification and intelligent user interaction for federated testbeds using Semantic Web technologies

Abstract In this paper we present the Testbed-as-a-Service with Ontology Repository (TaaSOR) platform that facilitates a community-based approach to building an infrastructure for the management and federation of heterogeneous networking testbeds in a bottom-up fashion. This platform provides semantically annotated services and experiments with automated transformation from other semi-structured data sources and a set of tools that enables a community of developers and experimenters to develop and modify ontologies which constitute the foundation of the platform. The platform also includes an ontology repository that enables persistence, publishing and management of the accumulated semantics in a transparent way based on ontologies and Linked Data represented in standardized formats. Positive feedback loops engineered in the platform lower the barrier for testbed operators for participation in a federation, reduce the barrier of entry for experimenters without expert domain knowledge, lower operational overhead for developers, and streamline operation for system administrators. As a proof of concept, the TaaSOR instance is applied to the ORBIT testbed and the corresponding OMF management framework to test the proposed platform in an environment used by a large number of researchers worldwide.

TaaSOR – Testbed-as-a-Service Ontology Repository

This demonstration is introducing Testbed-as-a-Service (TaaS) infrastructure that illustrates use of community based approach to building experimental ontologies and generation of supporting testbed resources applied to OMF based testbed. While this TaaS demo is initially primarily targeting virtualization and community collaboration, the final objective is to support domain specific experimental description languages and resource management in federated testbeds.

A RESTful technique for collaborative learning content transclusion by Wiki-style mashups

In this paper we propose a simple pragmatic technique, called fladget, for enabling end-users to mashup multimedia content within Wiki pages of their community peers. Since the fladget considers Wiki as a content as well as mashup repository service, Wiki RESTful API is proposed. The fladget extends functionality of existing plugin mechanism, so it can use rich-client technology for interaction with distributed multimedia content, but in a pragmatic Wiki-like manner. The presented concept is illustrated by a hypothetical Linked Active Learning Community example demonstrating how the presented mechanism can be used at the community interaction level.

The CoordSS experimentation framework ontologies

The CoordSS experimentation framework aims to provide a solid practical evidence that testbeds and other experimentation infrastructure provided by EU FP7 FLEX project is capable of supporting advanced experimentation with proposed use of LTE Advanced in unlicensed 5 GHz spectrum (LTE-U). By integrating technologies such as spectrum sensing, various coexistence mechanisms between legacy unlicensed technologies (primarily WiFi) and LTE, and a range of coordination protocols, the frameworks primary objective is to enable experimentation with SDR based LTE in unlicensed spectrum. The flexibility of such experimental environment is addressed by adoption of domain and system ontologies for formal representation of semantics of the problem. This paper deals with ontologies that are needed for the CoordSS experimentation framework implementation.

Collaborative Wiki Tagging

Wikis as well as collaborative tagging have been subject of very intense research and an active discussion topic in the so-called blogsphere. In this paper, we propose Collaborative Wiki Tagging based on the idea to exploit inherent semantics of the concept of link in a wiki. The low-level integration of wiki and collaborative tagging of web resources is expected to be effective in enterprise environments particularly in the personal and group knowledge management application area. We first introduce a conceptualization of Collaborative Wiki Tagging. Then, we propose a simple scheme for using one of the existing native wiki syntax to represent tagging data. Collaborative Wiki Tagging Portal Prototype, developed as a proof of concept, is used to give illustrative practical examples of the proposed approach and illustration of the user interface.

Semantic driven code generation for networking testbed experimentation

ABSTRACT This paper investigates how semantic technologies can be used for automatic code generation with a goal to provide easier to use and more efficient working environment to users of large complex infrastructures such as networking testbeds. A new Semantic Driven Code Generation (SecGENE) approach is proposed in the light of a testbed-driven experimentation. The target use-case scenario focuses on heterogeneous telecommunication systems and optimal spectrum usage. The SecGENE approach is examined on the example of semantic coordination process in an environment containing coordinated as well as uncoordinated WiFi networks. Abbreviations: CLOnt – Command Line ontology; ExFOnt – Experiment Flow Ontology; MSAO – Meta System Architecture ontology; NVF – Network Virtualization Function; OEDL – OMF Experiment Description Language; OMN- Open-Multinet ontology; RDFO – A standard W3C Resource Description Framework ontology; SOEDL– Semantic OEDL

Representation of spectrum sensing experimentation functionality for federated management and control

Spectrum sensing is one of the core functionalities of a true cognitive radio (CR) that supports operation over a broad range of frequencies and can autonomously adapt transmission parameters to the operating environment. There are several types of hardware ranging from sophisticated (i.e. Nutaq Radio420X FPGA mezzanine card) to low cost (i.e. WiSpy) that can be used to experiment with spectrum sensing. This hardware is available for use in several testbeds across the world (i.e. ORBIT, w-iLab.t, TWIST and LOG-a-TEC). Each testbed provides a specific mechanism to define, deploy and execute experiments making it difficult for an individual researcher to use more than one testbed. In this work we propose an information model for describing spectrum sensing functionality with the ultimate goal of developing and promoting a Common CR language that can describe the resources in existing GENI and FIRE testbeds.