With the availability of more and more cores within architectures the process of extracting implicit and explicit parallelism in applications to fully utilise these cores is becoming complex. Implicit parallelism extraction is performed through the inclusion of intelligent software and hardware sections of tool chains although these reach their theoretical limit rather quickly. Due to this the concept of a method of allowing explicit parallelism to be performed as fast a possible has been investigated. This method enables application developers to perform creation and synchronisation of parallel sections of an application at a finer-grained level than previously possible, resulting in smaller sections of code being executed in parallel while still reducing overall execution time. Alongside explicit parallelism, a concept of high level design of applications destined for multicore systems was also investigated. As systems are getting larger it is becoming more difficult to design and track the full life-cycle of development. One method used to ease this process is to use a graphical design process to visualise the high level designs of such systems. One drawback in graphical design is the explicit nature in which systems are required to be generated, this was investigated, and using concepts already in use in text based programming languages, the generation of platform-independent models which are able to be specialised to multiple hardware architectures was developed. The explicit parallelism was performed using hardware elements to perform thread management, this resulted in speed ups of over 13 times when compared to threading libraries executed in software on commercially available processors. This allowed applications with large data dependent sections to be parallelised in small sections within the code resulting in a decrease of overall execution time. The modelling concepts resulted in the saving of between 40-50% of the time and effort required to generate platform-specific models while only incurring an overhead of up to 15% the execution cycles of these models designed for specific architectures.

Multi-core and many-core systems are the norm in contemporary processor technology and are expected to remain so for the foreseeable future. Parallel programming is, thus, here to stay and programmers have to endorse it if they are to exploit such systems for their applications. Programs using parallel programming primitives like PThreads or OpenMP often exploit coarse-grain parallelism, because it offers a good trade-off between programming effort versus performance gain. Some parallel applications show limited or no scaling beyond a number of cores. Given the abundant number of cores expected in future many-cores, several cores would remain idle in such cases while execution performance stagnates. This thesis proposes using cores that do not contribute to performance improvement for running implicit fine-grain speculative threads. In particular, we present a many-core architecture and protocols that allow applications with coarse-grain explicit parallelism to further exploit implicit speculative parallelism within each thread. We show that complementing parallel programs with implicit speculative mechanisms offers significant performance improvements for a large and diverse set of parallel benchmarks. Implicit speculative parallelism frees the programmer from the additional effort to explicitly partition the work into finer and properly synchronized tasks. Our results show that, for a many-core comprising 128 cores supporting implicit speculative parallelism in clusters of 2 or 4 cores, performance improves on top of the highest scalability point by 44% on average for the 4-core cluster and by 31% on average for the 2-core cluster. We also show that this approach often leads to better performance and energy efficiency compared to existing alternatives such as Core Fusion and Turbo Boost. Moreover, we present a dynamic mechanism to choose the number of explicit and implicit threads, which performs within 6% of the static oracle selection of threads. To improve energy efficiency processors allow for Dynamic Voltage and Frequency Scaling (DVFS), which enables changing their performance and power consumption on-the-fly. We evaluate the amenability of the proposed explicit plus implicit threads scheme to traditional power management techniques for multithreaded applications and identify room for improvement. We thus augment prior schemes and introduce a novel multithreaded power management scheme that accounts for implicit threads and aims to minimize the Energy Delay2 product (ED2). Our scheme comprises two components: a “local” component that tries to adapt to the different program phases on a per explicit thread basis, taking into account implicit thread behavior, and a “global” component that augments the local components with information regarding inter-thread synchronization. Experimental results show a reduction of ED2 of 8% compared to having no power management, with an average reduction in power of 15% that comes at a minimal loss of performance of less than 3% on average.

The rapidly increasing number of cores available in multicore processors does not necessarily lead directly to a commensurate increase in performance: programs written in conventional languages, such as C, need careful restructuring, preferably automatically, before the benefits can be observed in improved run-times. Even then, much depends upon the intrinsic capacity of the original program for concurrent execution. Using software techniques to parallelize the sequential application can raise the level of gain from multicore systems. Parallel programming is not an easy job for the user, who has to deal with many issues such as dependencies, synchronization, load balancing, and race conditions. For this reason the role of automatically parallelizing compilers and techniques for the extraction of several threads from single-threaded programs, without programmer intervention, is becoming more important and may help to deliver better utilization of modern hardware. One parallelizing technique that has been shown to be an effective for the parallelization of applications that have irregular control flow and complex memory access patterns is Decoupled Software Pipeline (DSWP). This transformation partitions the loop body into a set of stages, ensuring that critical path dependencies are kept local to a stage. Each stage becomes a thread and data is passed between threads using inter-core communication. The success of DSWP depends on being able to extract the relatively fine-grain parallelism that is present in many applications. Another technique which offers potential gains in parallelizing general purpose applications is slicing. Program slicing transforms large programs into several smaller ones that execute independently, each consisting of only statements relevant to the computation of certain, socalled, (program) points. This dissertation explores the possibility of performance benefits arising from a secondary transformation of DSWP stages by slicing. To that end a new combination method called DSWP/Slice is presented. Our observation is that individual DSWP stages can be parallelized by slicing, leading to an improvement in performance of the longest duration DSWP stages. In particular, this approach can be applicable in cases where DOALL is not. In consequence better load balancing can be achieved between the DSWP stages. Moreover, we introduce an automatic implementation of the combination method using Low Level Virtual Machine (LLVM) compiler framework. This combination is particularly effective when the whole long stage comprises a function body. More than one slice extracted from a function body can speed up its execution time and also increases the scalability of DSWP. An evaluation of this technique on six programs with a range of dependence patterns leads to considerable performance gains on a core-i7 870 machine with 4-cores/8-threads. The results are obtained from an automatic implementation that shows the proposed method can give a factor of up to 1.8 speed up compared with the original sequential code.

As the number of embedded applications is increasing, the current strategy of several companies is to launch a new platform within short periods, to execute the application set more efficiently, with low energy consumption. However, for each new platform deployment, new tool chains must come along, with additional libraries, debuggers and compilers. This strategy implies in high hardware redesign costs, breaks binary compatibility and results in a high overhead in the software development process. Therefore, focusing on area savings, low energy consumption, binary compatibility maintenance and mainly software productivity improvement, we propose the exploitation of Custom Reconfigurable Arrays for Multiprocessor System (CReAMS). CReAMS is composed of multiple adaptive reconfigurable systems to efficiently explore Instruction and Thread Level Parallelism (ILP and TLP) at hardware level, in a totally transparent fashion. Conceived as homogeneous organization, CReAMS shows a reduction of 37% in energy-delay product (EDP) compared to an ordinary multiprocessing platform when assuming the same chip area. When a variety of processor with different capabilities on exploiting ILP are coupled in a single die, conceiving CReAMS as a heterogeneous organization, performance improvements of up to 57% and energy savings of up to 36% are showed in comparison with the homogenous platform. In addition, the efficiency of the adaptability provided by CReAMS is demonstrated in a comparison to a multiprocessing system composed of 4- issue Out-of-Order SparcV8 processors, 28% of performance improvements are shown considering a power budget scenario.

As the number of embedded applications is increasing, the current strategy of several companies is to launch a new platform within short periods, to execute the application set more efficiently, with low energy consumption. However, for each new platform deployment, new tool chains must come along, with additional libraries, debuggers and compilers. This strategy implies in high hardware redesign costs, breaks binary compatibility and results in a high overhead in the software development process. Therefore, focusing on area savings, low energy consumption, binary compatibility maintenance and mainly software productivity improvement, we propose the exploitation of Custom Reconfigurable Arrays for Multiprocessor System (CReAMS). CReAMS is composed of multiple adaptive reconfigurable systems to efficiently explore Instruction and Thread Level Parallelism (ILP and TLP) at hardware level, in a totally transparent fashion. Conceived as homogeneous organization, CReAMS shows a reduction of 37% in energy-delay product (EDP) compared to an ordinary multiprocessing platform when assuming the same chip area. When a variety of processor with different capabilities on exploiting ILP are coupled in a single die, conceiving CReAMS as a heterogeneous organization, performance improvements of up to 57% and energy savings of up to 36% are showed in comparison with the homogenous platform. In addition, the efficiency of the adaptability provided by CReAMS is demonstrated in a comparison to a multiprocessing system composed of 4- issue Out-of-Order SparcV8 processors, 28% of performance improvements are shown considering a power budget scenario.

As the number of embedded applications is increasing, the current strategy of several companies is to launch a new platform within short periods, to execute the application set more efficiently, with low energy consumption. However, for each new platform deployment, new tool chains must come along, with additional libraries, debuggers and compilers. This strategy implies in high hardware redesign costs, breaks binary compatibility and results in a high overhead in the software development process. Therefore, focusing on area savings, low energy consumption, binary compatibility maintenance and mainly software productivity improvement, we propose the exploitation of Custom Reconfigurable Arrays for Multiprocessor System (CReAMS). CReAMS is composed of multiple adaptive reconfigurable systems to efficiently explore Instruction and Thread Level Parallelism (ILP and TLP) at hardware level, in a totally transparent fashion. Conceived as homogeneous organization, CReAMS shows a reduction of 37% in energy-delay product (EDP) compared to an ordinary multiprocessing platform when assuming the same chip area. When a variety of processor with different capabilities on exploiting ILP are coupled in a single die, conceiving CReAMS as a heterogeneous organization, performance improvements of up to 57% and energy savings of up to 36% are showed in comparison with the homogenous platform. In addition, the efficiency of the adaptability provided by CReAMS is demonstrated in a comparison to a multiprocessing system composed of 4- issue Out-of-Order SparcV8 processors, 28% of performance improvements are shown considering a power budget scenario.

This dissertation describes detailed performance engineering and optimization of an unstructured computational aerodynamics software system with irregular memory accesses on various multi- and many-core emerging high performance computing scalable architectures, which are expected to be the building blocks of energy-austere exascale systems, and on which algorithmic- and architecture-oriented optimizations are essential for achieving worthy performance. We investigate several state-of-the-practice shared-memory optimization techniques applied to key kernels for the important problem class of unstructured meshes. We illustrate for a broad spectrum of emerging microprocessor architectures as representatives of the compute units in contemporary leading supercomputers, identifying and addressing performance challenges without compromising the floating-point numerics of the original code. While the linear algebraic kernels are bottlenecked by memory bandwidth for even modest numbers of hardware cores sharing a common address space, the edge-based loop kernels, which arise in the control volume discretization of the conservation law residuals and in the formation of the preconditioner for the Jacobian by finite-differencing the conservation law residuals, are compute-intensive and effectively exploit contemporary multi- and many-core processing hardware. We therefore employ low- and high-level algorithmic- and architecture-specific code optimizations and tuning in light of thread- and data-level parallelism, with a focus on strong thread scaling at the node-level. Our approaches are based upon novel multi-level hierarchical workload distribution mechanisms of data across different compute units (from the address space down to the registers) within every hardware core. We analyze the demonstrated aerodynamics application on specific computing architectures to develop certain performance metrics and models to bespeak the upper and lower bounds of the performance. We present significant full application speedup relative to the baseline code, on a succession of many-core processor architectures, i.e., Intel Xeon Phi Knights Corner (5.0x) and Knights Landing (2.9x). In addition, the performance of Knights Landing outperforms, at significantly lower power consumption, Intel Xeon Skylake with nearly twofold speedup. These optimizations are expected to be of value for many other unstructured mesh partial differential equation-based scientific applications as multi- and many- core architecture evolves.

Computer architects continue to push the limits of modern microprocessors. By using techniques such as out-of-order execution, branch prediction, and dynamic scheduling, designers have found ways to speed execution. However, growing architectural complexity has led to unsustained development and testing times. Shrinking feature sizes are causing increased wire resistances and signal propagation, thereby limiting a design's scalability. Indeed, the method of exploiting instruction-level parallelism (ILP) within applications is reaching a point of diminishing returns. One approach to the aforementioned challenges is the Single-Chip Message-Passing (SCMP) Parallel Computer, developed at Virginia Tech. SCMP is a unique, tiled architecture aimed at thread-level parallelism (TLP). Identical cores are replicated across the chip, and global wire traces have been eliminated. The nodes are connected via a 2-D grid network and each contains a local memory bank. This thesis presents the design and analysis of three high-performance applications for SCMP. The results show that the architecture proves itself as a formidable opponent to several current systems. / Master of Science

<p>QC 20160428</p>

The level of Thread-Level Parallelism (TLP), Instruction-Level Parallelism (ILP), and Memory-Level Parallelism (MLP) varies across programs and across program phases. Hence, every program requires different underlying core microarchitecture resources for high performance and/or energy efficiency. Current core microarchitectures are inefficient because they are fixed at design time and do not adapt to variable TLP, ILP, or MLP. I show that if a core microarchitecture can adapt to the variation in TLP, ILP, and MLP, significantly higher performance and/or energy efficiency can be achieved. I propose MorphCore, a low-overhead adaptive microarchitecture built from a traditional OOO core with small changes. MorphCore adapts to TLP by operating in two modes: (a) as a wide-width large-OOO-window core when TLP is low and ILP is high, and (b) as a high-performance low-energy highly-threaded in-order SMT core when TLP is high. MorphCore adapts to ILP and MLP by varying the superscalar width and the out-of-order (OOO) window size by operating in four modes: (1) as a wide-width large-OOO-window core, 2) as a wide-width medium-OOO-window core, 3) as a medium-width large-OOO-window core, and 4) as a medium-width medium-OOO-window core. My evaluation with single-thread and multi-thread benchmarks shows that when highest single-thread performance is desired, MorphCore achieves performance similar to a traditional out-of-order core. When energy efficiency is desired on single-thread programs, MorphCore reduces energy by up to 15% (on average 8%) over an out-of-order core. When high multi-thread performance is desired, MorphCore increases performance by 21% and reduces energy consumption by 20% over an out-of-order core. Thus, for multi-thread programs, MorphCore's energy efficiency is similar to highly-threaded throughput-optimized small and medium core architectures, and its performance is two-thirds of their potential. / text