Search

Deep learning (DL) methods have recently gained popularity. Training this class of models is, however, computing-intensive, and frequently GPUs are used to boost performance. Although the costs of GPU-based systems are gradually reducing due to the high demand, they are still prohibitive: in public clouds, GPU-powered virtual machines (VMs) time unit price is 5-8x higher than CPU-only VMs. While the cloud remains the most cost-effective and flexible deployment, operation costs can be reduced, in large settings, by rightsizing and sharing resources among multiple processes. This work addresses the online joint capacity planning and job scheduling with due dates problem and proposes alternative matheuristic solution methods. Our objective is to optimize operation costs by: i) rightsizing the VM capacities at each node, ii) partitioning the set of GPUs among multiple concurrent jobs on the same VM, and iii) determining a due-date-aware job schedule. The effectiveness of the proposed hierarchical approach, coupled with an appropriate Mixed Integer Linear Programming formulation, is validated against first-principle methods by relying on simulation. The experiments prove that the efficiency of GPU-based systems evaluated in terms of costs can be improved by 50-70%. Finally, scalability analyses show that the proposed approach enables to solve problem instances with up to 100 nodes in less than one minute on average, making it suitable for practical scenarios.

The transformation of nanoparticles (NPs) in physiological milieu is a dynamic phenomenon that is the subject of intense investigation. When introduced into the body, NPs can undergo a variety of changes, such as, protein adsorption, dissolution, agglomeration/aggregation, structural deformities and redox reactions. It is these changes that subsequently determine the uptake, bioavailability, translocation and fate of NPs, which ultimately determine their therapeutic efficiency, diagnostic efficacy or toxicity. This chapter will consider the colloidal interactions at the interface of NPs with the contents of biological milieu, the practical and theoretical considerations required to modify analytical and imaging techniques to detect and, if possible, quantify NPs in this complex environment, and the requirement for a highly interdisciplinary approach to understand the behaviour at the bio-nano interface.

INTRODUCTION: Patients with sepsis often receive large amounts of fluids and the presence of capillary leak, trauma or bleeding results in ongoing fluid resuscitation. This increases interstitial and intestinal edema and finally leads to intra-abdominal hypertension (IAH), which in turn impedes lymphatic drainage. Patients with IAH often develop secondary respiratory failure needing mechanical ventilation with high intrathoracic pressure or PEEP that might further alter lymphatic drainage. This review will try to convince the reader of the importance of the lymphatics in septic patients with IAH. METHODS: A Medline and PubMed literature search was performed using the terms "abdominal pressure", "lymphatic drainage" and "ascites formation". The references from these studies were searched for relevant articles that may have been missed in the primary search. These articles served as the basis for the recommendations below. RESULTS: Induction of sepsis with lesion of the capillary alveolar barrier results in an increased water gradient between the capillaries and the interstitium in the lungs. The drainage flow to the thoracic duct is initially increased in order to protect the Lung and maintain the pulmonary interstitium as dry as possible, however this results in increased intrathoracic pressure. Sepsis also increases the permeability of the capillaries in the splanchnic beds. In analogy to the lungs the lymphatic flow in the splanchnic areas increases together with the pressure inside as a physiological response in order to limit the increase in IAP. At a critical IAP level (around 20 cmH2O) the lymph flow starts to decrease and the splanchnic water content progressively increases. The lymph flow from the abdomen to the thorax is progressively decreased resulting in increased splanchnic water content and ascites formation. The presence of mechanical ventilation with high PEEP reduces the lymph drainage further which together with the increase in IAP decreases the lymphatic pressure gradient in the splanchnic regions, with a further increase in water content and IAP triggering a vicious cycle. CONCLUSION: Although often overlooked the role of lymphatic flow is complex but very important to determine not only the fluid balance in the lung but also in the peripheral organs. Different pathologies and treatments can markedly influence the pathophysiology of the lymphatics with dramatic effects on endorgan function.