We investigate the problem of resource allocation in heterogeneous networks of computational resources. We provide an explicit analytical solution for a situation where the computational environment can be described by M/M/l queueing theory. We illustrate the quality of our solution by comparing results with those obtained via a simple ad hoc resource allocation in large heterogeneous networks consisting of N = 104 nodes with computational resources distributed either uniformly in a given interval, or exponentially in R+. 相似文献
Fracture tests of graphite are known to exhibit sensitivity to stress state, such as a difference between their flexural and tensile strengths. Bi-axial tensile and flexural loading are representative of the stress states in some regions of graphite components in nuclear fission reactors, where loading develops from fast neutron irradiation-induced dimensional change and thermal strains. Study of the behaviour of the inherent defects that determine strength variability requires in situ observation of crack nucleation. To this end, digital image correlation can be used to monitor the evolution of displacement fields and hence the cracks on the surface of large samples whilst under load. In this study, a ring-on-ring flexural test setup was developed to apply equi-biaxial tensile stress to large disc specimens of graphite along with the conventional four-point-bend test. A 17% reduction in mean flexural strength was observed for the equi-biaxial loading, relative to uniaxial loading. DIC was used to characterise the observed fracture nuclei. Linear elastic fracture mechanics analysis was shown to be inadequate to explain the strength reduction. It is suggested that fictitious crack models, originally developed to simulate the behaviour of concrete structures, can be utilised to explain the behaviour. 相似文献
Key results are summarized of efforts to significantly reduce the near-field measurement time by utilizing one- or two-dimensional arrays of modulated scattering probes in lieu of the single probe ordinarily used in conventional near-field measurement techniques. Results of analytical, numerical, and experimental investigations show that the modulated-scattering technique (MST) using arrays of hundreds or even thousands of modulated scattering probes can be used to map the complex near-field of antennas or scatterers in a few seconds or minutes. The results also strongly indicate that classical (nonmodulated) receiving/transmitting arrays can be adapted for rapid near-field data collection. Major factors affecting the accuracy and speed of probe arrays for near-field measurement are delineated and discussed. Experimental results obtained using laboratory prototype MST systems are also presented and discussed 相似文献
A correction procedure is developed for the case of the continuous rotation of test antenna while performing fast near-field measurements. This general development is vital for the technique of modulated scattering which combines fast electronic scanning through the use of a probe array in one dimension and the rotation of the test antenna for the other dimension. A numerical simulation was performed to check the validity of the correction procedure. The correction procedure is not based on any interpolation scheme. 相似文献
This study investigated the behaviour and fate of Listeria monocytogenes at different ripening temperatures and NaCl concentrations in traditional Lighvan cheese. L. monocytogenes was added to raw sheep's milk. After producing the cheese, they were stored in 8%, 12% and 15% NaCl at 4, 9 and 14 °C. Sampling was performed for 150 days. Different temperature and NaCl concentrations had a significant effect on the survival of L. monocytogenes (P <0.001). The lowest growth and survival rates of L. monocytogenes were in 15% NaCl at 14 °C and 12% NaCl at 14 °C, respectively. Also, the highest growth and survival rates of the bacterium were in 8% NaCl at 4 °C. 相似文献
Sixteen patients with combined paresis and restriction of extraocular muscle(s) orbital fracture repair were studied before and after in order to determine the clinical features and management of such patients. All 16 patients showed limited ductions of the involved eye in the field of action of the entrapped, paretic muscle and of the antagonist muscle after orbital fracture. Single extraocular muscles (13 patients) and two extraocular muscles (three patients) were demonstrated adjacent to the fracture site on orbital computed tomography (CT). In three patients prior to orbital surgery, a deviation in primary position was present. After fracture repair with release of the entrapped muscle in all patients, evidence of paresis of the muscle was demonstrated by underaction in its field of action and overaction in the field of its antagonist. There was a resultant manifest tropia or phoria in the primary position. In seven patients, the paresis gradually improved with no tropia and little diplopia in the functional fields of gaze. Three patients had minimal deviations and required no further treatment. Six patients with significant deviations required prisms (three patients) or strabismus surgery (three patients). The latter three patients had two muscles involved. Results of this study demonstrate that the ophthalmologist must appropriately diagnose patients with paresis and restriction of an extraocular muscle and counsel them that "new" diplopia may occur after orbital fracture repair and that this diplopia may require additional therapy. 相似文献
In this article, fracture toughness behavior of tool steel was investigated using Acoustic Emission (AE) monitoring. Fracture toughness (KIC) values of a specific tool steel was determined by applying various approaches based on conventional AE parameters, such as Acoustic Emission Cumulative Count (AECC), Acoustic Emission Energy Rate (AEER), and the combination of mechanical characteristics and AE information called sentry function. The critical fracture toughness values during crack propagation were achieved by means of relationship between the integral of the sentry function and cumulative fracture toughness (KICUM). Specimens were selected from AISI D2 cold-work tool steel and were heat treated at four different tempering conditions (300, 450, 525, and 575?°C). The results achieved through AE approaches were then compared with a methodology proposed by compact specimen testing according to ASTM standard E399. It was concluded that AE information was an efficient method to investigate fracture characteristics. 相似文献
Chronic wounds are one of the most devastating complications of diabetes and are the leading cause of nontraumatic limb amputation. Despite the progress in identifying factors and promising in vitro results for the treatment of chronic wounds, their clinical translation is limited. Given the range of disruptive processes necessary for wound healing, different pharmacological agents are needed at different stages of tissue regeneration. This requires the development of wearable devices that can deliver agents to critical layers of the wound bed in a minimally invasive fashion. Here, for the first time, a programmable platform is engineered that is capable of actively delivering a variety of drugs with independent temporal profiles through miniaturized needles into deeper layers of the wound bed. The delivery of vascular endothelial growth factor (VEGF) through the miniaturized needle arrays demonstrates that, in addition to the selection of suitable therapeutics, the delivery method and their spatial distribution within the wound bed is equally important. Administration of VEGF to chronic dermal wounds of diabetic mice using the programmable platform shows a significant increase in wound closure, re‐epithelialization, angiogenesis, and hair growth when compared to standard topical delivery of therapeutics. 相似文献
Service availability plays a vital role on computer networks, against which Distributed Denial of Service (DDoS) attacks are an increasingly growing threat each year. Machine learning (ML) is a promising approach widely used for DDoS detection, which obtains satisfactory results for pre-known attacks. However, they are almost incapable of detecting unknown malicious traffic. This paper proposes a novel method combining both supervised and unsupervised algorithms. First, a clustering algorithm separates the anomalous traffic from the normal data using several flow-based features. Then, using certain statistical measures, a classification algorithm is used to label the clusters. Employing a big data processing framework, we evaluate the proposed method by training on the CICIDS2017 dataset and testing on a different set of attacks provided in the more up-to-date CICDDoS2019. The results demonstrate that the Positive Likelihood Ratio (LR+) of our method is approximately 198% higher than the ML classification algorithms.