Job scheduler for streaming applications in heterogeneous distributed processing systems

The Journal of Supercomputing - In this study, we investigated the problem of scheduling streaming applications on a heterogeneous cluster environment and, based on our previous work, developed the...     

A high performance algorithm for static task scheduling in heterogeneous distributed computing systems

Effective task scheduling is essential for obtaining high performance in heterogeneous distributed computing systems (HeDCSs). However, finding an effective task schedule in HeDCSs requires the consideration of both the heterogeneity of processors and high interprocessor communication overhead, which results from non-trivial data movement between tasks scheduled on different processors. In this paper, we present a new high-performance scheduling algorithm, called the longest dynamic critical path (LDCP) algorithm, for HeDCSs with a bounded number of processors. The LDCP algorithm is a list-based scheduling algorithm that uses a new attribute to efficiently select tasks for scheduling in HeDCSs. The efficient selection of tasks enables the LDCP algorithm to generate high-quality task schedules in a heterogeneous computing environment. The performance of the LDCP algorithm is compared to two of the best existing scheduling algorithms for HeDCSs: the HEFT and DLS algorithms. The comparison study shows that the LDCP algorithm outperforms the HEFT and DLS algorithms in terms of schedule length and speedup. Moreover, the improvement in performance obtained by the LDCP algorithm over the HEFT and DLS algorithms increases as the inter-task communication cost increases. Therefore, the LDCP algorithm provides a practical solution for scheduling parallel applications with high communication costs in HeDCSs.     

Reliable performance prediction for multigrid software on distributed memory systems

We propose a model for describing and predicting the parallel performance of a broad class of parallel numerical software on distributed memory architectures. The purpose of this model is to allow reliable predictions to be made for the performance of the software on large numbers of processors of a given parallel system, by only benchmarking the code on small numbers of processors. Having described the methods used, and emphasized the simplicity of their implementation, the approach is tested on a range of engineering software applications that are built upon the use of multigrid algorithms. Despite their simplicity, the models are demonstrated to provide both accurate and robust predictions across a range of different parallel architectures, partitioning strategies and multigrid codes. In particular, the effectiveness of the predictive methodology is shown for a practical engineering software implementation of an elastohydrodynamic lubrication solver.     

A UML-based quantitative framework for early prediction of resource usage and load in distributed real-time systems

This paper presents a quantitative framework for early prediction of resource usage and load in distributed real-time systems (DRTS). The prediction is based on an analysis of UML 2.0 sequence diagrams, augmented with timing information, to extract timed-control flow information. It is aimed at improving the early predictability of a DRTS by offering a systematic approach to predict, at the design phase, system behavior in each time instant during its execution. Since behavioral models such as sequence diagrams are available in early design phases of the software life cycle, the framework enables resource analysis at a stage when design decisions are still easy to change. Though we provide a general framework, we use network traffic as an example resource type to illustrate how the approach is applied. We also indicate how usage and load analysis of other types of resources (e.g., CPU and memory) can be performed in a similar fashion. A case study illustrates the feasibility of the approach.

Kassandra: A framework for distributed simulation of heterogeneous cooperating objects

To address the heterogeneity and scalability issues of simulating Cooperating Objects (COs) systems, we propose Kassandra, a conceptual framework for enabling distributed COs simulation by integrating existing simulation tools. Moreover, Kassandra exploits the communication middleware used by real-world COs as underlying communication mechanism for integrating Kassandra-enabled simulation tools. In this way, real-world COs can be included with simulated objects in a seamless way to perform more accurate system performance evaluation. Moreover, such a hardware-in-the-loop approach is not limited to pre-deployment performance analysis, and can offer possibilities to analyse performance at different phases of CO applications. The concept of Kassandra has been carried out in the EU PLANET project. In this paper, we introduce the Kassandra framework components and show their interactions at different phases for node deployments in PLANET use cases. The result demonstrates the applicability of Kassandra to facilitate the development of CO applications.     

A hybrid Branch-and-Bound and evolutionary approach for allocating strings of applications to heterogeneous distributed computing systems

Providing efficient workload management is an important issue for a large-scale heterogeneous distributed computing environment where a set of periodic applications is executed. The considered shipboard distributed system is expected to operate in an environment where the input workload is likely to change unpredictably, possibly invalidating a resource allocation that was based on the initial workload estimate. The tasks consist of multiple strings, each made up of an ordered sequence of applications. There is a quality of service (QoS) minimum throughput constraint that must be satisfied for each application in a string, and a maximum utilization constraint that must be satisfied on each of the hardware resources in the system. The challenge, therefore, is to efficiently and robustly manage both computation and communication resources in this unpredictable environment to achieve high performance while satisfying the imposed constraints. This work addresses the problem of finding a robust initial allocation of resources to strings of applications that is able to absorb some level of unknown input workload increase without rescheduling. The proposed hybrid two-stage method of finding a near-optimal allocation of resources incorporates two specially designed mapping techniques: (1) the Permutation Space Genitor-Based heuristic, and (2) the follow-up Branch-and-Bound heuristic based on an Integer Linear Programming (ILP) problem formulation. The performance of the proposed resource allocation method is evaluated under different simulation scenarios and compared to an iteratively computed upper bound.     

A uniform approach for programming distributed heterogeneous computing systems

Large-scale compute clusters of heterogeneous nodes equipped with multi-core CPUs and GPUs are getting increasingly popular in the scientific community. However, such systems require a combination of different programming paradigms making application development very challenging.In this article we introduce libWater, a library-based extension of the OpenCL programming model that simplifies the development of heterogeneous distributed applications. libWater consists of a simple interface, which is a transparent abstraction of the underlying distributed architecture, offering advanced features such as inter-context and inter-node device synchronization. It provides a runtime system which tracks dependency information enforced by event synchronization to dynamically build a DAG of commands, on which we automatically apply two optimizations: collective communication pattern detection and device-host-device copy removal.We assess libWater’s performance in three compute clusters available from the Vienna Scientific Cluster, the Barcelona Supercomputing Center and the University of Innsbruck, demonstrating improved performance and scaling with different test applications and configurations.     

Principles for designing data-/compute-intensive distributed applications and middleware systems for heterogeneous environments

The nature of distributed systems is constantly and steadily changing as the hardware and software landscape evolves. Porting applications and adapting existing middleware systems to ever changing computational platforms has become increasingly complex and expensive. Therefore, the design of applications, as well as the design of next generation middleware systems, must follow a set of guiding principles in order to insure long-term “survivability” without costly re-engineering. From our practical experience, the key determinants to success in this endeavor are adherence to the following principles: (1) Design for change; (2) Provide for storage subsystem I/O coordination; (3) Employ workload partitioning and load balancing techniques; (4) Employ caching; (5) Schedule the workload; and (6) Understand the workload. In order to support these principles, we have collected extensive experimental results comparing three middleware systems targeted at data- and compute-intensive applications implemented by our research group during the course of the last decade, on a single data- and compute-intensive application. The main contribution of this work is the analysis of a level playing field, where we discuss and quantify how adherence to these guiding principles impacts overall system throughput and response time.     

A framework for efficient regression tests on database applications

Regression testing is an important software maintenance activity to ensure the integrity of a software after modification. However, most methods and tools developed for software testing today do not work well for database applications; these tools only work well if applications are stateless or tests can be designed in such a way that they do not alter the state. To execute tests for database applications efficiently, the challenge is to control the state of the database during testing and to order the test runs such that expensive database reset operations that bring the database into the right state need to be executed as seldom as possible. This work devises a regression testing framework for database applications so that test runs can be executed in parallel. The goal is to achieve linear speed-up and/or exploit the available resources as well as possible. This problem is challenging because parallel testing needs to consider both load balancing and controlling the state of the database. Experimental results show that test run execution can achieve linear speed-up by using the proposed framework.     

A self-adaptable query allocation framework for distributed information systems

In large-scale distributed information systems, where participants are autonomous and have special interests for some queries, query allocation is a challenge. Much work in this context has focused on distributing queries among providers in a way that maximizes overall performance (typically throughput and response time). However, preserving the participants’ interests is also important. In this paper, we make the following contributions. First, we provide a model to define the participants’ perception of the system regarding their interests and propose measures to evaluate the quality of query allocation methods. Then, we propose a framework for query allocation called Satisfaction-based Query Load Balancing (SQLB, for short), which dynamically trades consumers’ interests for providers’ interests based on their satisfaction. Finally, we compare SQLB, through experimentation, with two important baseline query allocation methods, namely Capacity based and Mariposa-like. The results demonstrate that SQLB yields high efficiency while satisfying the participants’ interests and significantly outperforms the baseline methods. Work partially funded by ARA “Massive Data” of the French ministry of research (Respire project) and the European Strep Grid4All project.     

A relativistic temporal algebra for efficient design of distributed systems

Adequate methods for checking the specification and design of distributed systems must allow for reasoning about asynchronous activities; efficient methods must perform the reasoning in polynomial time. This paper lays the groundwork for such an efficient deductive system by providing a very general temporal relation algebra that can be used by constraint propagation techniques to perform the required reasoning. Major choices exist when selecting an appropriate temporal model: discrete/dense, linear/nonlinear, and point/interval. James Allen and others have indicated the possible atomic relations between two intervals for the dense-linear-interval model, while Anger, Ladkin, and Rodriguez have shown those needed for a dense-branching-interval model. Rodriguez and Anger further developed a dense-relativistic-interval model based on Lamport'sprecede andcan affect arrows, determining a large number of atomic relations. This paper shows that those same atomic relations are exactly the correct ones for intervals in dense relativistic space-time if intervals are taken as pairs of points (E _s ,E _f ) in space-time such that it is possible to move fromE _s toE _f at less than the speed of light. The relations are defined and named consistently with the earlier work of Rodriguez and Anger, and the relationship between the two models is pursued. The relevance of the results to the verification of distributed specifications and algorithms is discussed.     

A framework for adaptive collective communications for heterogeneous hierarchical computing systems

Collective communication operations are widely used in MPI applications and play an important role in their performance. However, the network heterogeneity inherent to grid environments represent a great challenge to develop efficient high performance computing applications. In this work we propose a generic framework based on communication models and adaptive techniques for dealing with collective communication patterns on grid platforms. Toward this goal, we address the hierarchical organization of the grid, selecting the most efficient communication algorithms at each network level. Our framework is also adaptive to grid load dynamics since it considers transient network characteristics for dividing the nodes into clusters. Our experiments with the broadcast operation on a real-grid setup indicate that an adaptive framework allows significant performance improvements on MPI collective communications.     

A framework for efficient and rapid development of cross-platform audio applications

In this article, we present CLAM, a C++ software framework, that offers a complete development and research platform for the audio and music domain. It offers an abstract model for audio systems and includes a repository of processing algorithms and data types as well as all the necessary tools for audio and control input/output. The framework offers tools that enable the exploitation of all these features to easily build cross-platform applications or rapid prototypes for media processing algorithms and systems. Furthermore, included ready-to-use applications can be used for tasks such as audio analysis/synthesis, plug-in development, feature extraction or metadata annotation. CLAM represents a step forward over other similar existing environments in the multimedia domain. Nevertheless, it also shares models and constructs with many of those. These commonalities are expressed in the form of a metamodel for multimedia processing systems and a design pattern language.     

Supervised ranking framework for relationship prediction in heterogeneous information networks

In recent years, relationship prediction in heterogeneous information networks (HINs) has become an active topic. The most essential part of this task is how to effectively represent and utilize the important three kinds of information hidden in connections of the network, namely local structure information (Local-info), global structure information (Global-info) and attribute information (Attr-info). Although all the information indicates different features of the network and influence relationship creation in a complementary way, existing approaches utilize them separately or in a partially combined way. In this article, a novel framework named Supervised Ranking framework (S-Rank) is proposed to tackle this issue. To avoid the class imbalance problem, in S-Rank framework we treat the relationship prediction problem as a ranking task and divide it into three phases. Firstly, a Supervised PageRank strategy (SPR) is proposed to rank the candidate nodes according to Global-info and Attr-info. Secondly, a Meta Path-based Ranking method (MPR) utilizing Local-info is proposed to rank the candidate nodes based on their meta path-based features. Finally, the two ranking scores are linearly integrated into the final ranking result which combines all the Attr-info, Global-info and Local-info together. Experiments on DBLP data demonstrate that the proposed S-Rank framework can effectively take advantage of all the three kinds of information for relationship prediction over HINs and outperforms other well-known baseline approaches.     

A performance model for analysis of heterogeneous multi-cluster systems

This paper addresses the problem of performance modeling for large-scale heterogeneous distributed systems with emphases on multi-cluster computing systems. Since the overall performance of distributed systems is often depends on the effectiveness of its communication network, the study of the interconnection networks for these systems is very important. Performance modeling is required to avoid poorly chosen components and architectures as well as discovering a serious shortfall during system testing just prior to deployment time. However, the multiplicity of components and associated complexity make performance analysis of distributed computing systems a challenging task. To this end, we present an analytical performance model for the interconnection networks of heterogeneous multi-cluster systems. The analysis is based on a parametric family of fat-trees, the m-port n-tree, and a deterministic routing algorithm, which is proposed in this paper. The model is validated through comprehensive simulation, which demonstrated that the proposed model exhibits a good degree of accuracy for various system organizations and under different working conditions.     

A framework for modular analysis and exploration of heterogeneous embedded systems

The increasing complexity of heterogeneous systems-on-chip, SoC, and distributed embedded systems makes system optimization and exploration a challenging task. Ideally, a designer would try all possible system configurations and choose the best one regarding specific system requirements. Unfortunately, such an approach is not possible because of the tremendous number of design parameters with sophisticated effects on system properties. Consequently, good search techniques are needed to find design alternatives that best meet constraints and cost criteria. In this paper, we present a compositional design space exploration framework for system optimization and exploration using SymTA/S, a software tool for formal performance analysis. In contrast to many previous approaches pursuing closed automated exploration strategies over large sets of system parameters, our approach allows the designer to effectively control the exploration process to quickly find good design alternatives. An important aspect and key novelty of our approach is system optimization with traffic shaping.     

Reliability-aware scheduling strategy for heterogeneous distributed computing systems

Heterogeneous computing systems are promising computing platforms, since single parallel architecture based systems may not be sufficient to exploit the available parallelism with the running applications. In some cases, heterogeneous distributed computing (HDC) systems can achieve higher performance with lower cost than single-machine supersystems. However, in HDC systems, processors and networks are not failure free and any kind of failure may be critical to the running applications. One way of dealing with such failures is to employ a reliable scheduling algorithm. Unfortunately, most existing scheduling algorithms for precedence constrained tasks in HDC systems do not adequately consider reliability requirements of inter-dependent tasks. In this paper, we design a reliability-driven scheduling architecture that can effectively measure system reliability, based on an optimal reliability communication path search algorithm, and then we introduce reliability priority rank (RRank) to estimate the task’s priority by considering reliability overheads. Furthermore, based on directed acyclic graph (DAG) we propose a reliability-aware scheduling algorithm for precedence constrained tasks, which can achieve high quality of reliability for applications. The comparison studies, based on both randomly generated graphs and the graphs of some real applications, show that our scheduling algorithm outperforms the existing scheduling algorithms in terms of makespan, scheduling length ratio, and reliability. At the same time, the improvement gained by our algorithm increases as the data communication among tasks increases.     

Adaptive load sharing in heterogeneous distributed systems

In this paper, we study the performance characteristics of simple load sharing algorithms for heterogeneous distributed systems. We assume that nonnegligible delays are encountered in transferring jobs from one node to another. We analyze the effects of these delays on the performance of two threshold-based algorithms called Forward and Reverse. We formulate queuing theoretic models for each of the algorithms operating in heterogeneous systems under the assumption that the job arrival process at each node in Poisson and the service times and job transfer times are exponentially distributed. The models are solved using the Matrix-Geometric solution technique. These models are used to study the effects of different parameters and algorithm variations on the mean job response time: e.g., the effects of varying the thresholds, the impact of changing the probe limit, the impact of biasing the probing, and the optimal response times over a large range of loads and delays. Wherever relevant, the results of the models are compared with the M/M/ 1 model, representing no load balancing (hereafter referred to as NLB), and the M/M/K model, which is an achievable lower bound (hereafter referred to as LB).     

Object orientation in heterogeneous distributed computing systems

The basic properties of object orientation and their application to heterogeneous, autonomous, and distributed system to increase interoperability ar examined. It is argued that object-oriented distributed computing is a natural step forward from client-server systems. To support this claim, the differing levels of object-oriented support already found in commercially available distributed systems-in particular, the distributed computing environment of the open software foundation and the Cronus system of Bolt Beranek, Newman (BBN)-are discussed. Emerging object-oriented systems and standards are described, focusing on the convergence toward a least-common-denominator approach to object-oriented distributed computing embodied by the object management group's common object request broker architecture     

Methodology evaluation framework for dynamic evolution in composition-based distributed applications

Dynamic evolution can be used to upgrade distributed applications without shutdown and restart as a way of improving service levels while minimising the loss of business revenue caused by the downtime. An evaluation framework assessing the level of support offered by existing methodologies in composition-based application (e.g. component-based and service-oriented) development is proposed. It was developed by an analysis of the literature and existing methodologies together with a refinement based on a survey of experienced practitioners and researchers. The use of the framework is demonstrated by applying it to twelve methodologies to assess their support for dynamic evolution.