Parallel programming in biomedical signal processing

Dissertação para obtenção do Grau de Mestre em Engenharia Biomédica

A system’s approach to cache hierarchy-aware decomposition of data-parallel computations

Dissertação para obtenção do Grau de Mestre em Engenharia Informática

Software for Explicitly Parallel Memory-Centric Processor Architecture

Advances in computer memory technology justify research towards new and different views on computer organization. This paper proposes a novel memory-centric computing architecture with the goal to merge memory and processing elements in order to provide better conditions for parallelization and performance. The paper introduces the architectural concepts and afterwards shows the design and implementation of a corresponding assembler and simulator.

Memory-based scheduling of scientific computing clusters

This study looks at how increased memory utilisation affects throughput and energy consumption in scientific computing, especially in high-energy physics. Our aim is to minimise energy consumed by a set of jobs without increasing the processing time. The earlier tests indicated that, especially in data analysis, throughput can increase over 100% and energy consumption decrease 50% by processing multiple jobs in parallel per CPU core. Since jobs are heterogeneous, it is not possible to find an optimum value for the number of parallel jobs. A better solution is based on memory utilisation, but finding an optimum memory threshold is not straightforward. Therefore, a fuzzy logic-based algorithm was developed that can dynamically adapt the memory threshold based on the overall load. In this way, it is possible to keep memory consumption stable with different workloads while achieving significantly higher throughput and energy-efficiency than using a traditional fixed number of jobs or fixed memory threshold approaches.

FastEpistasis: a high performance computing solution for quantitative trait epistasis.

Motivation: Genome-wide association studies have become widely used tools to study effects of genetic variants on complex diseases. While it is of great interest to extend existing analysis methods by considering interaction effects between pairs of loci, the large number of possible tests presents a significant computational challenge. The number of computations is further multiplied in the study of gene expression quantitative trait mapping, in which tests are performed for thousands of gene phenotypes simultaneously. Results: We present FastEpistasis, an efficient parallel solution extending the PLINK epistasis module, designed to test for epistasis effects when analyzing continuous phenotypes. Our results show that the algorithm scales with the number of processors and offers a reduction in computation time when several phenotypes are analyzed simultaneously. FastEpistasis is capable of testing the association of a continuous trait with all single nucleotide polymorphism ( SNP) pairs from 500 000 SNPs, totaling 125 billion tests, in a population of 5000 individuals in 29, 4 or 0.5 days using 8, 64 or 512 processors.

Coscheduling techniques and monitoring tools for non-dedicated cluster computing

Our efforts are directed towards the understanding of the coscheduling mechanism in a NOW system when a parallel job is executed jointly with local workloads, balancing parallel performance against the local interactive response. Explicit and implicit coscheduling techniques in a PVM-Linux NOW (or cluster) have been implemented. Furthermore, dynamic coscheduling remains an open question when parallel jobs are executed in a non-dedicated Cluster. A basis model for dynamic coscheduling in Cluster systems is presented in this paper. Also, one dynamic coscheduling algorithm for this model is proposed. The applicability of this algorithm has been proved and its performance analyzed by simulation. Finally, a new tool (named Monito) for monitoring the different queues of messages in such an environments is presented. The main aim of implementing this facility is to provide a mean of capturing the bottlenecks and overheads of the communication system in a PVM-Linux cluster.

Memristive Computing

Memristive computing refers to the utilization of the memristor, the fourth fundamental passive circuit element, in computational tasks. The existence of the memristor was theoretically predicted in 1971 by Leon O. Chua, but experimentally validated only in 2008 by HP Labs. A memristor is essentially a nonvolatile nanoscale programmable resistor — indeed, memory resistor — whose resistance, or memristance to be precise, is changed by applying a voltage across, or current through, the device. Memristive computing is a new area of research, and many of its fundamental questions still remain open. For example, it is yet unclear which applications would benefit the most from the inherent nonlinear dynamics of memristors. In any case, these dynamics should be exploited to allow memristors to perform computation in a natural way instead of attempting to emulate existing technologies such as CMOS logic. Examples of such methods of computation presented in this thesis are memristive stateful logic operations, memristive multiplication based on the translinear principle, and the exploitation of nonlinear dynamics to construct chaotic memristive circuits. This thesis considers memristive computing at various levels of abstraction. The first part of the thesis analyses the physical properties and the current-voltage behaviour of a single device. The middle part presents memristor programming methods, and describes microcircuits for logic and analog operations. The final chapters discuss memristive computing in largescale applications. In particular, cellular neural networks, and associative memory architectures are proposed as applications that significantly benefit from memristive implementation. The work presents several new results on memristor modeling and programming, memristive logic, analog arithmetic operations on memristors, and applications of memristors. The main conclusion of this thesis is that memristive computing will be advantageous in large-scale, highly parallel mixed-mode processing architectures. This can be justified by the following two arguments. First, since processing can be performed directly within memristive memory architectures, the required circuitry, processing time, and possibly also power consumption can be reduced compared to a conventional CMOS implementation. Second, intrachip communication can be naturally implemented by a memristive crossbar structure.

Video transcoding in a distributed cloud computing environment

Video transcoding refers to the process of converting a digital video from one format into another format. It is a compute-intensive operation. Therefore, transcoding of a large number of simultaneous video streams requires a large amount of computing resources. Moreover, to handle di erent load conditions in a cost-e cient manner, the video transcoding service should be dynamically scalable. Infrastructure as a Service Clouds currently offer computing resources, such as virtual machines, under the pay-per-use business model. Thus the IaaS Clouds can be leveraged to provide a coste cient, dynamically scalable video transcoding service. To use computing resources e ciently in a cloud computing environment, cost-e cient virtual machine provisioning is required to avoid overutilization and under-utilization of virtual machines. This thesis presents proactive virtual machine resource allocation and de-allocation algorithms for video transcoding in cloud computing. Since users' requests for videos may change at di erent times, a check is required to see if the current computing resources are adequate for the video requests. Therefore, the work on admission control is also provided. In addition to admission control, temporal resolution reduction is used to avoid jitters in a video. Furthermore, in a cloud computing environment such as Amazon EC2, the computing resources are more expensive as compared with the storage resources. Therefore, to avoid repetition of transcoding operations, a transcoded video needs to be stored for a certain time. To store all videos for the same amount of time is also not cost-e cient because popular transcoded videos have high access rate while unpopular transcoded videos are rarely accessed. This thesis provides a cost-e cient computation and storage trade-o strategy, which stores videos in the video repository as long as it is cost-e cient to store them. This thesis also proposes video segmentation strategies for bit rate reduction and spatial resolution reduction video transcoding. The evaluation of proposed strategies is performed using a message passing interface based video transcoder, which uses a coarse-grain parallel processing approach where video is segmented at group of pictures level.

Implementing Green IT approach for transferring Big Data over Parallel Data Link

The whole research of the current Master Thesis project is related to Big Data transfer over Parallel Data Link and my main objective is to assist the Saint-Petersburg National Research University ITMO research team to accomplish this project and apply Green IT methods for the data transfer system. The goal of the team is to transfer Big Data by using parallel data links with SDN Openflow approach. My task as a team member was to compare existing data transfer applications in case to verify which results the highest data transfer speed in which occasions and explain the reasons. In the context of this thesis work a comparison between 5 different utilities was done, which including Fast Data Transfer (FDT), BBCP, BBFTP, GridFTP, and FTS3. A number of scripts where developed which consist of creating random binary data to be incompressible to have fair comparison between utilities, execute the Utilities with specified parameters, create log files, results, system parameters, and plot graphs to compare the results. Transferring such an enormous variety of data can take a long time, and hence, the necessity appears to reduce the energy consumption to make them greener. In the context of Green IT approach, our team used Cloud Computing infrastructure called OpenStack. It’s more efficient to allocated specific amount of hardware resources to test different scenarios rather than using the whole resources from our testbed. Testing our implementation with OpenStack infrastructure results that the virtual channel does not consist of any traffic and we can achieve the highest possible throughput. After receiving the final results we are in place to identify which utilities produce faster data transfer in different scenarios with specific TCP parameters and we can use them in real network data links.

A User-Centric Perspective on Parallel Programming with Focus on OpenMP

The process of developing software that takes advantage of multiple processors is commonly referred to as parallel programming. For various reasons, this process is much harder than the sequential case. For decades, parallel programming has been a problem for a small niche only: engineers working on parallelizing mostly numerical applications in High Performance Computing. This has changed with the advent of multi-core processors in mainstream computer architectures. Parallel programming in our days becomes a problem for a much larger group of developers. The main objective of this thesis was to find ways to make parallel programming easier for them. Different aims were identified in order to reach the objective: research the state of the art of parallel programming today, improve the education of software developers about the topic, and provide programmers with powerful abstractions to make their work easier. To reach these aims, several key steps were taken. To start with, a survey was conducted among parallel programmers to find out about the state of the art. More than 250 people participated, yielding results about the parallel programming systems and languages in use, as well as about common problems with these systems. Furthermore, a study was conducted in university classes on parallel programming. It resulted in a list of frequently made mistakes that were analyzed and used to create a programmers' checklist to avoid them in the future. For programmers' education, an online resource was setup to collect experiences and knowledge in the field of parallel programming - called the Parawiki. Another key step in this direction was the creation of the Thinking Parallel weblog, where more than 50.000 readers to date have read essays on the topic. For the third aim (powerful abstractions), it was decided to concentrate on one parallel programming system: OpenMP. Its ease of use and high level of abstraction were the most important reasons for this decision. Two different research directions were pursued. The first one resulted in a parallel library called AthenaMP. It contains so-called generic components, derived from design patterns for parallel programming. These include functionality to enhance the locks provided by OpenMP, to perform operations on large amounts of data (data-parallel programming), and to enable the implementation of irregular algorithms using task pools. AthenaMP itself serves a triple role: the components are well-documented and can be used directly in programs, it enables developers to study the source code and learn from it, and it is possible for compiler writers to use it as a testing ground for their OpenMP compilers. The second research direction was targeted at changing the OpenMP specification to make the system more powerful. The main contributions here were a proposal to enable thread-cancellation and a proposal to avoid busy waiting. Both were implemented in a research compiler, shown to be useful in example applications, and proposed to the OpenMP Language Committee.

A parallel genetic algorithm for the Steiner Problem in Networks

This paper presents a parallel genetic algorithm to the Steiner Problem in Networks. Several previous papers have proposed the adoption of GAs and others metaheuristics to solve the SPN demonstrating the validity of their approaches. This work differs from them for two main reasons: the dimension and the characteristics of the networks adopted in the experiments and the aim from which it has been originated. The reason that aimed this work was namely to build a comparison term for validating deterministic and computationally inexpensive algorithms which can be used in practical engineering applications, such as the multicast transmission in the Internet. On the other hand, the large dimensions of our sample networks require the adoption of a parallel implementation of the Steiner GA, which is able to deal with such large problem instances.

Building collaborative environments for advanced computing

Synchronous collaborative systems allow geographically distributed users to form a virtual work environment enabling cooperation between peers and enriching the human interaction. The technology facilitating this interaction has been studied for several years and various solutions can be found at present. In this paper, we discuss our experiences with one such widely adopted technology, namely the Access Grid [1]. We describe our experiences with using this technology, identify key problem areas and propose our solution to tackle these issues appropriately. Moreover, we propose the integration of Access Grid with an Application Sharing tool, developed by the authors. Our approach allows these integrated tools to utilise the enhanced features provided by our underlying dynamic transport layer.

Hierarchical composite regular parallel architecture

The design space of emerging heterogenous multi-core architectures with re-configurability element makes it feasible to design mixed fine-grained and coarse-grained parallel architectures. This paper presents a hierarchical composite array design which extends the curret design space of regular array design by combining a sequence of transformations. This technique is applied to derive a new design of a pipelined parallel regular array with different dataflow between phases of computation.

Parallel Monte Carlo approach for integration of the rendering equation

This paper is addressed to the numerical solving of the rendering equation in realistic image creation. The rendering equation is integral equation describing the light propagation in a scene accordingly to a given illumination model. The used illumination model determines the kernel of the equation under consideration. Nowadays, widely used are the Monte Carlo methods for solving the rendering equation in order to create photorealistic images. In this work we consider the Monte Carlo solving of the rendering equation in the context of the parallel sampling scheme for hemisphere. Our aim is to apply this sampling scheme to stratified Monte Carlo integration method for parallel solving of the rendering equation. The domain for integration of the rendering equation is a hemisphere. We divide the hemispherical domain into a number of equal sub-domains of orthogonal spherical triangles. This domain partitioning allows to solve the rendering equation in parallel. It is known that the Neumann series represent the solution of the integral equation as a infinity sum of integrals. We approximate this sum with a desired truncation error (systematic error) receiving the fixed number of iteration. Then the rendering equation is solved iteratively using Monte Carlo approach. At each iteration we solve multi-dimensional integrals using uniform hemisphere partitioning scheme. An estimate of the rate of convergence is obtained using the stratified Monte Carlo method. This domain partitioning allows easy parallel realization and leads to convergence improvement of the Monte Carlo method. The high performance and Grid computing of the corresponding Monte Carlo scheme are discussed.