Use of shared memory in the context of embedded multi-core processors: exploration of the technology and its limits

Modern embedded systems embrace many-core shared-memory designs. Due to constrained power and area budgets, most of them feature software-managed scratchpad memories instead of data caches to increase the data locality. It is therefore programmers’ responsibility to explicitly manage the memory transfers, and this make programming these platform cumbersome. Moreover, complex modern applications must be adequately parallelized before they can the parallel potential of the platform into actual performance. To support this, programming languages were proposed, which work at a high level of abstraction, and rely on a runtime whose cost hinders performance, especially in embedded systems, where resources and power budget are constrained. This dissertation explores the applicability of the shared-memory paradigm on modern many-core systems, focusing on the ease-of-programming. It focuses on OpenMP, the de-facto standard for shared memory programming. In a first part, the cost of algorithms for synchronization and data partitioning are analyzed, and they are adapted to modern embedded many-cores. Then, the original design of an OpenMP runtime library is presented, which supports complex forms of parallelism such as multi-level and irregular parallelism. In the second part of the thesis, the focus is on heterogeneous systems, where hardware accelerators are coupled to (many-)cores to implement key functional kernels with orders-of-magnitude of speedup and energy efficiency compared to the “pure software” version. However, three main issues rise, namely i) platform design complexity, ii) architectural scalability and iii) programmability. To tackle them, a template for a generic hardware processing unit (HWPU) is proposed, which share the memory banks with cores, and the template for a scalable architecture is shown, which integrates them through the shared-memory system. Then, a full software stack and toolchain are developed to support platform design and to let programmers exploiting the accelerators of the platform. The OpenMP frontend is extended to interact with it.

Parallel Architectures for Many-Core Systems-On-Chip in Deep Sub-Micron Technology

Despite the several issues faced in the past, the evolutionary trend of silicon has kept its constant pace. Today an ever increasing number of cores is integrated onto the same die. Unfortunately, the extraordinary performance achievable by the many-core paradigm is limited by several factors. Memory bandwidth limitation, combined with inefficient synchronization mechanisms, can severely overcome the potential computation capabilities. Moreover, the huge HW/SW design space requires accurate and flexible tools to perform architectural explorations and validation of design choices. In this thesis we focus on the aforementioned aspects: a flexible and accurate Virtual Platform has been developed, targeting a reference many-core architecture. Such tool has been used to perform architectural explorations, focusing on instruction caching architecture and hybrid HW/SW synchronization mechanism. Beside architectural implications, another issue of embedded systems is considered: energy efficiency. Near Threshold Computing is a key research area in the Ultra-Low-Power domain, as it promises a tenfold improvement in energy efficiency compared to super-threshold operation and it mitigates thermal bottlenecks. The physical implications of modern deep sub-micron technology are severely limiting performance and reliability of modern designs. Reliability becomes a major obstacle when operating in NTC, especially memory operation becomes unreliable and can compromise system correctness. In the present work a novel hybrid memory architecture is devised to overcome reliability issues and at the same time improve energy efficiency by means of aggressive voltage scaling when allowed by workload requirements. Variability is another great drawback of near-threshold operation. The greatly increased sensitivity to threshold voltage variations in today a major concern for electronic devices. We introduce a variation-tolerant extension of the baseline many-core architecture. By means of micro-architectural knobs and a lightweight runtime control unit, the baseline architecture becomes dynamically tolerant to variations.

Regulating FinTech: The perspectives of law, economics, and technology

FinTech (financial technology, ‘‘FinTech’’) is a double-edged sword as it brings both benefits and risks. This study appraised FinTech’s technological nature that brings changes in complexity in modern financial markets to identify the information deficits and its undesirable outcomes. Besides, as FinTech is still developing, the information regarding, for instance, whether and how to apply regulation may be insufficient for both regulators and those regulated. More one-size-fits-all regulation might accordingly be adopted, thereby resulting in the adverse selection. Through the lens of both law and economics and law and technology, this study suggested AFR (adaptive financial regulation, ‘‘AFR’’) of FinTech to solve the underlying pacing issue. AFR is dynamic, enabling regulatory adjustments and learning. Exploring and collecting information through experiments and learning from experiments are the core of AFR. FinTech regulatory sandboxes epitomize AFR. This study chose Taiwan as a case study. This study found several barriers to adaptive and effective FinTech regulation. Unduly emphasizing consumer protection and the innovation entry criterion by improperly imposing limits on the entry into sandboxes, ignoring post-sandbox mechanisms, and relying on detailed, specific and prescriptive rules to formulate sandboxes are examples. To solve these barriers, this study proposed several solutions by looking into the experiences in other jurisdictions and analyzing. First, striking a balance between encouraging innovation and ensuring financial stability and consumer protection is indispensable. Second, entry to sandboxes should be facilitated by improving the selection criteria. Third, adhering to realizing regulatory adjustment and learning to adapt regulation to technology, this study argued that systematic post-sandbox mechanisms should be established. Fourth, this study recommended “more principles-based sandboxes”. Principles rather than rules should be the base on which sandboxes or FinTech regulation are established. Having principles could provide more flexibility, being easier to adjust and adapt, and better at avoiding.

New senthesis and derivatizations of peptidomimetics and peptide-conjugates for theranostic applications

In the contest of a modern green chemistry approach, we firstly tried to substituent the classic peptide synthesis approach with the use of N-carboxyanhydrides in the presence of Hydroxyapatite, a high biocompatible inorganic base. Despite the great results, further developments are necessary for a daily use in laboratory and for our research, we decided to proceed with solid phase or liquid phase synthesis. In the first chapter, the treatment of pain with the use of opioids is introduced. The abuse and misuse of these kind of potent analgesics, led to the necessity of developing new drugs with less side effects. Starting from a previous study, where the introduction of a lactam-like structure in the place of the proline of Endomorphine1, switched the selectivity from MOR to KOR, we designed and synthetized three different libraries by placing a different trans inducer element to gain the desired selectivity and activity forcing the structure to adopt a linear rather than folded position. In the second chapter, we focused on lactate dehydrogenase, an enzyme overexpressed when the cells in hypoxia conditions, like in a tumour mass, need to produce energy through the transformation of pyruvate into lactate. We synthetized different cyclic peptidomimetics, designed to be inhibitors, as powerful tool to contrast cancer cells growing. Biological assays produced satisfactory preliminary results, but further studies are necessary for a definitive output. Finally in the last chapter, the cancer treatment problem is also approached through the design of nanoparticles, able to deliver drugs with efficacy and selectivity. We firstly synthetized silica core nanoparticles, built with toxic peptide sequences conjugated through click chemistry with Pluronic acid and then, in collaboration with Miriam Royo’s research group, we synthetized multivalent platforms for used drugs for the treatment of advanced colorectal cancer.