Data-parallel computation in parallel and distributed environments

The past few decades have seen a considerable increase in the number of parallel and distributed systems. With the development of more complex applications, the need for more powerful systems has emerged and various parallel and distributed environments have been designed and implemented. Each of the environments, including hardware and software, has unique strengths and weaknesses. There is no single parallel environment that can be identified as the best environment for all applications with respect to hardware and software properties. The main goal of this thesis is to provide a novel way of performing data-parallel computation in parallel and distributed environments by utilizing the best characteristics of difference aspects of parallel computing. For the purpose of this thesis, three aspects of parallel computing were identified and studied. First, three parallel environments (shared memory, distributed memory, and a network of workstations) are evaluated to quantify theirsuitability for different parallel applications. Due to the parallel and distributed nature of the environments, networks connecting the processors in these environments were investigated with respect to their performance characteristics. Second, scheduling algorithms are studied in order to make them more efficient and effective. A concept of application-specific information scheduling is introduced. The application- specific information is data about the workload extractedfrom an application, which is provided to a scheduling algorithm. Three scheduling algorithms are enhanced to utilize the application-specific information to further refine their scheduling properties. A more accurate description of the workload is especially important in cases where the workunits are heterogeneous and the parallel environment is heterogeneous and/or non-dedicated. The results obtained show that the additional information regarding the workload has a positive impact on the performance of applications. Third, a programming paradigm for networks of symmetric multiprocessor (SMP) workstations is introduced. The MPIT programming paradigm incorporates the Message Passing Interface (MPI) with threads to provide a methodology to write parallel applications that efficiently utilize the available resources and minimize the overhead. The MPIT allows for communication and computation to overlap by deploying a dedicated thread for communication. Furthermore, the programming paradigm implements an application-specific scheduling algorithm. The scheduling algorithm is executed by the communication thread. Thus, the scheduling does not affect the execution of the parallel application. Performance results achieved from the MPIT show that considerable improvements over conventional MPI applications are achieved.

Sulautetun moniprosessorijärjestelmän suunnittelu sähkömoottorien ohjaus- ja valvontatarpeisiin

Universal Converter (UNICON) –projektin osana suunniteltiin sähkömoottorikäyttöjen ohjaukseen ja mittaukseen soveltuva digitaaliseen signaaliprosessoriin (DSP) pohjautuva sulautettu järjestelmä. Riittävän laskentatehon varmistamiseksi päädyttiin käyttämään moniprosessorijärjestelmää. Prosessorijärjestelmässä käytettävää DSP-piiriä valittaessa valintaperusteina olivat piirien tarjoama prosessointiteho ja moniprosessorituki. Analog Devices:n SHARC-sarjan DSP-piirit täyttivät parhaiten asetetut vaatimukset: Ne tarjoavat tehokkaan käskykannan lisäksi suuren sisäisen muistin ja sisäänrakennetun moniprosessorituen. Järjestelmän mittalaiteluonteisuudesta johtuen keskeinen suunnitteluparametri oli luoda nopeat tiedonsiirtoyhteydet mittausantureilta DSP-järjestelmään. Tämä toteutettiin käyttäen ohjelmointavia FPGA-logiikkapiirejä digitaalimuotoisen mittausdatan vastaanotossa ja esikäsittelyssä. Tiedonsiirtoyhteys PC-tietokoneelle toteutettiin käyttäen erityistä liityntäkorttia DSP-järjestelmän ja PC-tietokoneen välillä. Liityntäkortin päätehtävänä on puskuroida siirrettävä data. Järjestelyllä estetään PC-tietokoneen vaikutus DSP-järjestelmän toimintaan, jotta kyetään takaamaan järjestelmän reaaliaikainen toiminta kaikissa olosuhteissa.

Exploring Adaptive Implementation of On-Chip Networks

As technology geometries have shrunk to the deep submicron regime, the communication delay and power consumption of global interconnections in high performance Multi- Processor Systems-on-Chip (MPSoCs) are becoming a major bottleneck. The Network-on- Chip (NoC) architecture paradigm, based on a modular packet-switched mechanism, can address many of the on-chip communication issues such as performance limitations of long interconnects and integration of large number of Processing Elements (PEs) on a chip. The choice of routing protocol and NoC structure can have a significant impact on performance and power consumption in on-chip networks. In addition, building a high performance, area and energy efficient on-chip network for multicore architectures requires a novel on-chip router allowing a larger network to be integrated on a single die with reduced power consumption. On top of that, network interfaces are employed to decouple computation resources from communication resources, to provide the synchronization between them, and to achieve backward compatibility with existing IP cores. Three adaptive routing algorithms are presented as a part of this thesis. The first presented routing protocol is a congestion-aware adaptive routing algorithm for 2D mesh NoCs which does not support multicast (one-to-many) traffic while the other two protocols are adaptive routing models supporting both unicast (one-to-one) and multicast traffic. A streamlined on-chip router architecture is also presented for avoiding congested areas in 2D mesh NoCs via employing efficient input and output selection. The output selection utilizes an adaptive routing algorithm based on the congestion condition of neighboring routers while the input selection allows packets to be serviced from each input port according to its congestion level. Moreover, in order to increase memory parallelism and bring compatibility with existing IP cores in network-based multiprocessor architectures, adaptive network interface architectures are presented to use multiple SDRAMs which can be accessed simultaneously. In addition, a smart memory controller is integrated in the adaptive network interface to improve the memory utilization and reduce both memory and network latencies. Three Dimensional Integrated Circuits (3D ICs) have been emerging as a viable candidate to achieve better performance and package density as compared to traditional 2D ICs. In addition, combining the benefits of 3D IC and NoC schemes provides a significant performance gain for 3D architectures. In recent years, inter-layer communication across multiple stacked layers (vertical channel) has attracted a lot of interest. In this thesis, a novel adaptive pipeline bus structure is proposed for inter-layer communication to improve the performance by reducing the delay and complexity of traditional bus arbitration. In addition, two mesh-based topologies for 3D architectures are also introduced to mitigate the inter-layer footprint and power dissipation on each layer with a small performance penalty.

On Dynamic Monitoring Methods for Networks-on-Chip

Rapid ongoing evolution of multiprocessors will lead to systems with hundreds of processing cores integrated in a single chip. An emerging challenge is the implementation of reliable and efficient interconnection between these cores as well as other components in the systems. Network-on-Chip is an interconnection approach which is intended to solve the performance bottleneck caused by traditional, poorly scalable communication structures such as buses. However, a large on-chip network involves issues related to congestion problems and system control, for instance. Additionally, faults can cause problems in multiprocessor systems. These faults can be transient faults, permanent manufacturing faults, or they can appear due to aging. To solve the emerging traffic management, controllability issues and to maintain system operation regardless of faults a monitoring system is needed. The monitoring system should be dynamically applicable to various purposes and it should fully cover the system under observation. In a large multiprocessor the distances between components can be relatively long. Therefore, the system should be designed so that the amount of energy-inefficient long-distance communication is minimized. This thesis presents a dynamically clustered distributed monitoring structure. The monitoring is distributed so that no centralized control is required for basic tasks such as traffic management and task mapping. To enable extensive analysis of different Network-on-Chip architectures, an in-house SystemC based simulation environment was implemented. It allows transaction level analysis without time consuming circuit level implementations during early design phases of novel architectures and features. The presented analysis shows that the dynamically clustered monitoring structure can be efficiently utilized for traffic management in faulty and congested Network-on-Chip-based multiprocessor systems. The monitoring structure can be also successfully applied for task mapping purposes. Furthermore, the analysis shows that the presented in-house simulation environment is flexible and practical tool for extensive Network-on-Chip architecture analysis.

Design Space Exploration for MPSoC Architectures

Multiprocessor system-on-chip (MPSoC) designs utilize the available technology and communication architectures to meet the requirements of the upcoming applications. In MPSoC, the communication platform is both the key enabler, as well as the key differentiator for realizing efficient MPSoCs. It provides product differentiation to meet a diverse, multi-dimensional set of design constraints, including performance, power, energy, reconfigurability, scalability, cost, reliability and time-to-market. The communication resources of a single interconnection platform cannot be fully utilized by all kind of applications, such as the availability of higher communication bandwidth for computation but not data intensive applications is often unfeasible in the practical implementation. This thesis aims to perform the architecture-level design space exploration towards efficient and scalable resource utilization for MPSoC communication architecture. In order to meet the performance requirements within the design constraints, careful selection of MPSoC communication platform, resource aware partitioning and mapping of the application play important role. To enhance the utilization of communication resources, variety of techniques such as resource sharing, multicast to avoid re-transmission of identical data, and adaptive routing can be used. For implementation, these techniques should be customized according to the platform architecture. To address the resource utilization of MPSoC communication platforms, variety of architectures with different design parameters and performance levels, namely Segmented bus (SegBus), Network-on-Chip (NoC) and Three-Dimensional NoC (3D-NoC), are selected. Average packet latency and power consumption are the evaluation parameters for the proposed techniques. In conventional computing architectures, fault on a component makes the connected fault-free components inoperative. Resource sharing approach can utilize the fault-free components to retain the system performance by reducing the impact of faults. Design space exploration also guides to narrow down the selection of MPSoC architecture, which can meet the performance requirements with design constraints.

A Model-Based Development and Verification Framework for Distributed System-on-Chip Architecture

The capabilities and thus, design complexity of VLSI-based embedded systems have increased tremendously in recent years, riding the wave of Moore’s law. The time-to-market requirements are also shrinking, imposing challenges to the designers, which in turn, seek to adopt new design methods to increase their productivity. As an answer to these new pressures, modern day systems have moved towards on-chip multiprocessing technologies. New architectures have emerged in on-chip multiprocessing in order to utilize the tremendous advances of fabrication technology. Platform-based design is a possible solution in addressing these challenges. The principle behind the approach is to separate the functionality of an application from the organization and communication architecture of hardware platform at several levels of abstraction. The existing design methodologies pertaining to platform-based design approach don’t provide full automation at every level of the design processes, and sometimes, the co-design of platform-based systems lead to sub-optimal systems. In addition, the design productivity gap in multiprocessor systems remain a key challenge due to existing design methodologies. This thesis addresses the aforementioned challenges and discusses the creation of a development framework for a platform-based system design, in the context of the SegBus platform - a distributed communication architecture. This research aims to provide automated procedures for platform design and application mapping. Structural verification support is also featured thus ensuring correct-by-design platforms. The solution is based on a model-based process. Both the platform and the application are modeled using the Unified Modeling Language. This thesis develops a Domain Specific Language to support platform modeling based on a corresponding UML profile. Object Constraint Language constraints are used to support structurally correct platform construction. An emulator is thus introduced to allow as much as possible accurate performance estimation of the solution, at high abstraction levels. VHDL code is automatically generated, in the form of “snippets” to be employed in the arbiter modules of the platform, as required by the application. The resulting framework is applied in building an actual design solution for an MP3 stereo audio decoder application.