A study of the concurrent execution of parallel and sequential applications on a non-dedicated cluster

Computers of a non-dedicated cluster are often idle (users attend meetings, have lunch or coffee breaks) or lightly loaded (users carry out simple computations to support problem solving activities). These underutilised computers can be employed to execute parallel applications. Thus, these computers can be shared by parallel and sequential applications, which could lead to the improvement of their execution performance. However, there is a lack of experimental study showing the applications’ performance and the system utilization of executing parallel and sequential applications concurrently and concurrent execution of multiple parallel applications on a non-dedicated cluster. Here we present the result of an experimental study into load balancing based scheduling of mixtures of NAS Parallel Benchmarks and BYTE sequential applications on a very low cost non-dedicated cluster. This study showed that the proposed sharing provided performance boost as compared to the execution of the parallel load in isolation on a reduced number of computers and better cluster utilization. The results of this research were used not only to validate other researchers’ result generated by simulation but also to support our research mission of widening the use of non-dedicated clusters. Our promising results obtained could promote further research studies to convince universities, business and industry, which require a large amount of computing resources, to run parallel applications on their already owned non-dedicated clusters.

Platform for reliable computing on clusters using group communications

Shared clusters represent an excellent platform for the execution of parallel applications given their low price/performance ratio and the presence of cluster infrastructure in many organisations. The focus of recent research efforts are on parallelism management, transport and efficient access to resources, and making clusters easy to use. In this thesis, we examine reliable parallel computing on clusters. The aim of this research is to demonstrate the feasibility of developing an operating system facility providing transport fault tolerance using existing, enhanced and newly built operating system services for supporting parallel applications. In particular, we use existing process duplication and process migration services, and synthesise a group communications facility for use in a transparent checkpointing facility. This research is carried out using the methods of experimental computer science. To provide a foundation for the synthesis of the group communications and checkpointing facilities, we survey and review related work in both fields. For group communications, we examine the V Distributed System, the x-kernel and Psync, the ISIS Toolkit, and Horus. We identify a need for services that consider the placement of processes on computers in the cluster. For Checkpointing, we examine Manetho, KeyKOS, libckpt, and Diskless Checkpointing. We observe the use of remote computer memories for storing checkpoints, and the use of copy-on-write mechanisms to reduce the time to create a checkpoint of a process. We propose a group communications facility providing two sets of services: user-oriented services and system-oriented services. User-oriented services provide transparency and target application. System-oriented services supplement the user-oriented services for supporting other operating systems services and do not provide transparency. Additional flexibility is achieved by providing delivery and ordering semantics independently. An operating system facility providing transparent checkpointing is synthesised using coordinated checkpointing. To ensure a consistent set of checkpoints are generated by the facility, instead of blindly blocking the processes of a parallel application, only non-deterministic events are blocked. This allows the processes of the parallel application to continue execution during the checkpoint operation. Checkpoints are created by adapting process duplication mechanisms, and checkpoint data is transferred to remote computer memories and disk for storage using the mechanisms of process migration. The services of the group communications facility are used to coordinate the checkpoint operation, and to transport checkpoint data to remote computer memories and disk. Both the group communications facility and the checkpointing facility have been implemented in the GENESIS cluster operating system and provide proof-of-concept. GENESIS uses a microkernel and client-server based operating system architecture, and is demonstrated to provide an appropriate environment for the development of these facilities. We design a number of experiments to test the performance of both the group communications facility and checkpointing facility, and to provide proof-of-performance. We present our approach to testing, the challenges raised in testing the facilities, and how we overcome them. For group communications, we examine the performance of a number of delivery semantics. Good speed-ups are observed and system-oriented group communication services are shown to provide significant performance advantages over user-oriented semantics in the presence of packet loss. For checkpointing, we examine the scalability of the facility given different levels of resource usage and a variable number of computers. Low overheads are observed for checkpointing a parallel application. It is made clear by this research that the microkernel and client-server based cluster operating system provide an ideal environment for the development of a high performance group communications facility and a transparent checkpointing facility for generating a platform for reliable parallel computing on clusters.

Some aspects of adaptive cluster sampling

The adaptive cluster sampling (ACS) procedure is difficult to apply if some of the networks appearing in the sample are large. To deal with such large networks, a two-stage adaptive cluster sampling (TACS) procedure and an adjusted two-stage adaptive cluster sampling (ATACS) procedure are discussed.

Parallel computing on clusters and enterprise grids : practice and experience

We assert that companies can make more money and research institutions can improve their performance if inexpensive clusters and enterprise grids are exploited. In this paper, we have demonstrated that our claim is valid by showing the study of how programming environments, tools and middleware could be used for the execution of parallel and sequential applications, multiple parallel applications executing simultaneously on a non-dedicated cluster, and parallel applications on an enterprise grid and that the execution performance was improved. For this purpose an execution environment, and parallel and sequential benchmark applications selected for, and used in, the experiments were characterised.

Toward higher level abstractions for cloud computing

The general idea of cloud computing is offering computational resources as scalable, on demand services accessible over the internet. However, this new realisation of on demand computing is hindered by the amount of user involvement. Currently, high level abstractions in cloud computing only exist in the form of services. In particular, there is no effective means to publish, discover and finally use required services in clouds. In response, we propose a user level abstraction on top of already available cloud abstraction layers, present its concept using the resources via web services (RVWS) framework and demonstrate its feasibility by simplifying the exposure and use of clusters, a commonly used resource in cloud computing.

Modelo de virtualização distribuída aplicado ao gerenciamento e replicação de cluster multiuso

Pós-graduação em Ciência da Computação - IBILCE

Approcci innovativi per future simulazioni HPC verso l'exascale computing con applicazioni alla fisica dei materiali

In questo lavoro di tesi sono state evidenziate alcune problematiche relative alle macchine exascale (sistemi che sviluppano un exaflops di Potenza di calcolo) e all'evoluzione dei software che saranno eseguiti su questi sistemi, prendendo in esame principalmente la necessità del loro sviluppo, in quanto indispensabili per lo studio di problemi scientifici e tecnologici di più grandi dimensioni, con particolare attenzione alla Material Science, che è uno dei campi che ha avuto maggiori sviluppi grazie all'utilizzo di supercomputer, ed ad uno dei codici HPC più utilizzati in questo contesto: Quantum ESPRESSO. Dal punto di vista del software sono state presentate le prime misure di efficienza energetica su architettura ibrida grazie al prototipo di cluster EURORA sul software Quantum ESPRESSO. Queste misure sono le prime ad essere state pubblicate nel contesto software per la Material Science e serviranno come baseline per future ottimizzazioni basate sull'efficienza energetica. Nelle macchine exascale infatti uno dei requisiti per l'accesso sarà la capacità di essere energeticamente efficiente, così come oggi è un requisito la scalabilità del codice. Un altro aspetto molto importante, riguardante le macchine exascale, è la riduzione del numero di comunicazioni che riduce il costo energetico dell'algoritmo parallelo, poiché in questi nuovi sistemi costerà di più, da un punto di vista energetico, spostare i dati che calcolarli. Per tale motivo in questo lavoro sono state esposte una strategia, e la relativa implementazione, per aumentare la località dei dati in uno degli algoritmi più dispendiosi, dal punto di vista computazionale, in Quantum ESPRESSO: Fast Fourier Transform (FFT). Per portare i software attuali su una macchina exascale bisogna iniziare a testare la robustezza di tali software e i loro workflow su test case che stressino al massimo le macchine attualmente a disposizione. In questa tesi per testare il flusso di lavoro di Quantum ESPRESSO e WanT, un software per calcolo di trasporto, è stato caratterizzato un sistema scientificamente rilevante costituito da un cristallo di PDI - FCN2 che viene utilizzato per la costruzione di transistor organici OFET. Infine è stato simulato un dispositivo ideale costituito da due elettrodi in oro con al centro una singola molecola organica.

Implementazione di una infrastruttura basata su Kafka e Storm per il Mobile Cloud computing

Lo scopo dell'elaborato di tesi è l'analisi, progettazione e sviluppo di un prototipo di una infrastruttura cloud in grado di gestire un grande flusso di eventi generati da dispositivi mobili. Questi utilizzano informazioni come la posizione assunta e il valore dei sensori locali di cui possono essere equipaggiati al fine di realizzare il proprio funzionamento. Le informazioni così ottenute vengono trasmesse in modo da ottenere una rete di device in grado di acquisire autonomamente informazioni sull'ambiente ed auto-organizzarsi. La costruzione di tale struttura si colloca in un più ampio ambito di ricerca che punta a integrare metodi per la comunicazione ravvicinata con il cloud al fine di permettere la comunicazione tra dispositivi vicini in qualsiasi situazione che si potrebbe presentare in una situazione reale. A definire le specifiche della infrastruttura e quindi a impersonare il ruolo di committente è stato il relatore, Prof. Mirko Viroli, mentre lo sviluppo è stato portato avanti da me e dal correlatore, Ing. Pietro Brunetti. Visti gli studi precedenti riguardanti il cloud computing nell'area dei sistemi complessi distribuiti, Brunetti ha dato il maggiore contributo nella fase di analisi del problema e di progettazione mentre la parte riguardante la effettiva gestione degli eventi, le computazioni in cloud e lo storage dei dati è stata maggiormente affrontata da me. In particolare mi sono occupato dello studio e della implementazione del backend computazionale, basato sulla tecnologia Apache Storm, della componente di storage dei dati, basata su Neo4j, e della costruzione di un pannello di visualizzazione basato su AJAX e Linkurious. A questo va aggiunto lo studio su Apache Kafka, utilizzato come tecnologia per realizzare la comunicazione asincrona ad alte performance tra le componenti. Si è reso necessario costruire un simulatore al fine di condurre i test per verificare il funzionamento della infrastruttura prototipale e per saggiarne l'effettiva scalabilità, considerato il potenziale numero di dispositivi da sostenere che può andare dalle decine alle migliaia. La sfida più importante riguarda la gestione della vicinanza tra dispositivi e la possibilità di scalare la computazione su più macchine. Per questo motivo è stato necessario far uso di tecnologie per l'esecuzione delle operazioni di memorizzazione, calcolo e trasmissione dei dati in grado di essere eseguite su un cluster e garantire una accettabile fault-tolerancy. Da questo punto di vista i lavori che hanno portato alla costruzione della infrastruttura sono risultati essere un'ottima occasione per prendere familiarità con tecnologie prima sconosciute. Quasi tutte le tecnologie utilizzate fanno parte dell'ecosistema Apache e, come esposto all'interno della tesi, stanno ricevendo una grande attenzione da importanti realtà proprio in questo periodo, specialmente Apache Storm e Kafka. Il software prodotto per la costruzione della infrastruttura è completamente sviluppato in Java a cui si aggiunge la componente web di visualizzazione sviluppata in Javascript.

Development of a high performance parallel computing platform and its use in the study of nanostructures : clusters, sheets and tubes .

Small clusters of gallium oxide, technologically important high temperature ceramic, together with interaction of nucleic acid bases with graphene and small-diameter carbon nanotube are focus of first principles calculations in this work. A high performance parallel computing platform is also developed to perform these calculations at Michigan Tech. First principles calculations are based on density functional theory employing either local density or gradient-corrected approximation together with plane wave and gaussian basis sets. The bulk Ga2O3 is known to be a very good candidate for fabricating electronic devices that operate at high temperatures. To explore the properties of Ga2O3 at nonoscale, we have performed a systematic theoretical study on the small polyatomic gallium oxide clusters. The calculated results find that all lowest energy isomers of GamOn clusters are dominated by the Ga-O bonds over the metal-metal or the oxygen-oxygen bonds. Analysis of atomic charges suggest the clusters to be highly ionic similar to the case of bulk Ga2O3. In the study of sequential oxidation of these slusters starting from Ga2O, it is found that the most stable isomers display up to four different backbones of constituent atoms. Furthermore, the predicted configuration of the ground state of Ga2O is recently confirmed by the experimental result of Neumark's group. Guided by the results of calculations the study of gallium oxide clusters, performance related challenge of computational simulations, of producing high performance computers/platforms, has been addressed. Several engineering aspects were thoroughly studied during the design, development and implementation of the high performance parallel computing platform, rama, at Michigan Tech. In an attempt to stay true to the principles of Beowulf revolutioni, the rama cluster was extensively customized to make it easy to understand, and use - for administrators as well as end-users. Following the results of benchmark calculations and to keep up with the complexity of systems under study, rama has been expanded to a total of sixty four processors. Interest in the non-covalent intereaction of DNA with carbon nanotubes has steadily increased during past several years. This hybrid system, at the junction of the biological regime and the nanomaterials world, possesses features which make it very attractive for a wide range of applicatioins. Using the in-house computational power available, we have studied details of the interaction between nucleic acid bases with graphene sheet as well as high-curvature small-diameter carbon nanotube. The calculated trend in the binding energies strongly suggests that the polarizability of the base molecules determines the interaction strength of the nucleic acid bases with graphene. When comparing the results obtained here for physisorption on the small diameter nanotube considered with those from the study on graphene, it is observed that the interaction strength of nucleic acid bases is smaller for the tube. Thus, these results show that the effect of introducing curvature is to reduce the binding energy. The binding energies for the two extreme cases of negligible curvature (i.e. flat graphene sheet) and of very high curvature (i.e. small diameter nanotube) may be considered as upper and lower bounds. This finding represents an important step towards a better understanding of experimentally observed sequence-dependent interaction of DNA with Carbon nanotubes.

Optical quantum computation using cluster states

We propose an approach to optical quantum computation in which a deterministic entangling quantum gate may be performed using, on average, a few hundred coherently interacting optical elements (beam splitters, phase shifters, single photon sources, and photodetectors with feedforward). This scheme combines ideas from the optical quantum computing proposal of Knill, Laflamme, and Milburn [Nature (London) 409, 46 (2001)], and the abstract cluster-state model of quantum computation proposed by Raussendorf and Briegel [Phys. Rev. Lett. 86, 5188 (2001)].

Cluster-state quantum computation

This article is a short introduction to and review of the cluster-state model of quantum computation, in which coherent quantum information processing is accomplished via a sequence of single-qubit measurements applied to a fixed quantum state known as a cluster state. We also discuss a few novel properties of the model, including a proof that the cluster state cannot occur as the exact ground state of any naturally occurring physical system, and a proof that measurements on any quantum state which is linearly prepared in one dimension can be efficiently simulated on a classical computer, and thus are not candidates for use as a substrate for quantum computation.

Noise thresholds for optical cluster-state quantum computation

In this paper we do a detailed numerical investigation of the fault-tolerant threshold for optical cluster-state quantum computation. Our noise model allows both photon loss and depolarizing noise, as a general proxy for all types of local noise other than photon loss noise. We obtain a threshold region of allowed pairs of values for the two types of noise. Roughly speaking, our results show that scalable optical quantum computing is possible in the combined presence of both noise types, provided that the loss probability is less than 3 X 10(-3) and the depolarization probability is less than 10(-4). Our fault-tolerant protocol involves a number of innovations, including a method for syndrome extraction known as telecorrection, whereby repeated syndrome measurements are guaranteed to agree. This paper is an extended version of Dawson.

Efficiently computing weighted proximity relationships in spatial databases

Spatial data mining recently emerges from a number of real applications, such as real-estate marketing, urban planning, weather forecasting, medical image analysis, road traffic accident analysis, etc. It demands for efficient solutions for many new, expensive, and complicated problems. In this paper, we investigate the problem of evaluating the top k distinguished “features” for a “cluster” based on weighted proximity relationships between the cluster and features. We measure proximity in an average fashion to address possible nonuniform data distribution in a cluster. Combining a standard multi-step paradigm with new lower and upper proximity bounds, we presented an efficient algorithm to solve the problem. The algorithm is implemented in several different modes. Our experiment results not only give a comparison among them but also illustrate the efficiency of the algorithm.

An Adaptation of the Hoshen-Kopelman Cluster Counting Algorithm for Honeycomb Networks

We develop a simplified implementation of the Hoshen-Kopelman cluster counting algorithm adapted for honeycomb networks. In our implementation of the algorithm we assume that all nodes in the network are occupied and links between nodes can be intact or broken. The algorithm counts how many clusters there are in the network and determines which nodes belong to each cluster. The network information is stored into two sets of data. The first one is related to the connectivity of the nodes and the second one to the state of links. The algorithm finds all clusters in only one scan across the network and thereafter cluster relabeling operates on a vector whose size is much smaller than the size of the network. Counting the number of clusters of each size, the algorithm determines the cluster size probability distribution from which the mean cluster size parameter can be estimated. Although our implementation of the Hoshen-Kopelman algorithm works only for networks with a honeycomb (hexagonal) structure, it can be easily changed to be applied for networks with arbitrary connectivity between the nodes (triangular, square, etc.). The proposed adaptation of the Hoshen-Kopelman cluster counting algorithm is applied to studying the thermal degradation of a graphene-like honeycomb membrane by means of Molecular Dynamics simulation with a Langevin thermostat. ACM Computing Classification System (1998): F.2.2, I.5.3.

Benchmarking of distributed computing engines spark and GraphLab for big data analytics

In this paper we evaluate and compare two representativeand popular distributed processing engines for large scalebig data analytics, Spark and graph based engine GraphLab. Wedesign a benchmark suite including representative algorithmsand datasets to compare the performances of the computingengines, from performance aspects of running time, memory andCPU usage, network and I/O overhead. The benchmark suite istested on both local computer cluster and virtual machines oncloud. By varying the number of computers and memory weexamine the scalability of the computing engines with increasingcomputing resources (such as CPU and memory). We also runcross-evaluation of generic and graph based analytic algorithmsover graph processing and generic platforms to identify thepotential performance degradation if only one processing engineis available. It is observed that both computing engines showgood scalability with increase of computing resources. WhileGraphLab largely outperforms Spark for graph algorithms, ithas close running time performance as Spark for non-graphalgorithms. Additionally the running time with Spark for graphalgorithms over cloud virtual machines is observed to increaseby almost 100% compared to over local computer clusters.