Contrary to the traditional belief that at least two energy-storage elements (capacitor and/or inductor) and a locally-active nonlinearity are needed to build an electronic oscillator, this paper presents an oscillator made with only two circuit elements, namely, a memristor and a battery. This simplest of all physical oscillators also serves as a textbook example for explaining the intimate relationship between the super-critical Hopf bifurcation phenomenon and the edge of chaos.
An increasing amount of analytics is performed on data that is procured in a real-time fashion to make real-time decisions. Such tasks include simple reporting on streams to sophisticated model building. However, the practicality of such analyses are impeded in several domains because they are faced with a fundamental trade-off between data collection latency and analysis accuracy. In this paper, we study this trade-off in the context of a specific domain, Cellular Radio Access Networks (RAN). Our choice of this domain is influenced by its commonalities with several other domains that produce real-time data, our access to a large live dataset, and their real-time nature and dimensionality which makes it a natural fit for a popular analysis technique, machine learning (ML). We find that the latency accuracy trade-off can be resolved using two broad, general techniques: intelligent data grouping and task formulations that leverage domain characteristics. Based on this, we present CellScope, a system that addresses this challenge by applying a domain specific formulation and application of Multi-task Learning (MTL) to RAN performance analysis. It achieves this goal using three techniques: feature engineering to transform raw data into effective features, a PCA inspired similarity metric to group data from geographically nearby base stations sharing performance commonalities, and a hybrid online-offline model for efficient model updates. Our evaluation of CellScope shows that its accuracy improvements over direct application of ML range from 2.5x to 4.4x while reducing the model update overhead by up to 4.8x. We have also used CellScope to analyze a live LTE consisting of over 2 million subscribers for a period of over 10 months, where it uncovered several problems and insights, some of them previously unknown.
The critical voltage for passivity breakdown on metals and steels in contact with halide-containing solutions is often observed to depend on anion size. In the case of iron and iron-base alloys, and for other metals and alloys (e.g., nickel-base and aluminum alloys), the greatest susceptibility toward passivity breakdown occurs with chloride ion. In this paper, an explanation of anion size effects is offered based upon the Point Defect Model. Briefly, the size effect is attributed to the competing effects of the energy and entropy (Gibbs energy) changes for expansion of a surface oxygen vacancy to a size that can accommodate the halide and the Gibbs energy of halide and possibly vacancy hydration/dehydration. The theory successfully accounts for the effect of anion size on the critical breakdown voltage of Fe-13Cr stainless steel.
A mutant of Arabidopsis thaliana (L.) Heynh. which lacks leaf starch was isolated by screening for plants which did not stain with iodine. The starchless phenotype, confirmed by quantitative enzymic analysis, is caused by a single recessive nuclear mutation which results in a deficiency of the chloroplast isozyme of phosphoglucomutase. When grown in a 12-h photoperiod, leaves of the wild-type accumulated substantial amounts of starch but lower levels of soluble sugars. Under these conditions, the mutant accumulated relatively high levels of soluble sugars. Rates of growth and net photosynthesis of the mutant and wild-type were indistinguishable when the plants were grown in constant illumination. However, in a short photoperiod, the growth of the mutant was severely impaired, the rate of photosynthesis was depressed relative to the wild-type, and the rate of dark respiration, which was high following the onset of darkness, exhibited an uncharacteristic decay throughout the dark period. The altered control of respiration by the mutant, which may be related to the relatively high levels of soluble carbohydrate that accumulate in the leaf and stem tissue, is believed to be partially responsible for the low growth rate of the mutant in short days. The depressed photosynthetic capacity of the mutant may also reflect a metabolic adaptation to the accumulation of high levels of soluble carbohydrate which mimics the effects of alterations in source/sink ratio. The activities of sucrose phosphate synthase and acid invertase are significantly higher in the mutant than in the wild-type whereas ADP-glucose pyrophosphorylase activity is lower. This suggests that the activities of these enzymes may be modulated in response to metabolite concentrations or flux through the pathways.
A two-layer continuous-time cellular neural network for finding the Radon transform of a binary image is presented. The functionality of this cellular neural network follows from the functionality of the connected component detector cellular neural network.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">></ETX>
Sufficient conditions for local and global asymptotic stability of equilibria of some general classes of neural networks are presented. In the event that the interconnection matrix is block diagonally stable it is shown that the equilibrium is globally asymptotically stable if the cells are dissipative at the equilibrium. For a special class of networks the conditions of dissipativity are reduced to more readily-tested conditions of passivity. Equilibria are shown to be asymptotically stable essentially if the cells are locally passive.
Many applications do not constrain the destinations of their network transfers. New opportunities emerge when such transfers contribute a large amount of network bytes. By choosing the endpoints to avoid congested links, completion times of these transfers as well as that of others without similar flexibility can be improved. In this paper, we focus on leveraging the flexibility in replica placement during writes to cluster file systems (CFSes), which account for almost half of all cross-rack traffic in data-intensive clusters. The replicas of a CFS write can be placed in any subset of machines as long as they are in multiple fault domains and ensure a balanced use of storage throughout the cluster. We study CFS interactions with the cluster network, analyze optimizations for replica placement, and propose Sinbad -- a system that identifies imbalance and adapts replica destinations to navigate around congested links. Experiments on EC2 and trace-driven simulations show that block writes complete 1.3X (respectively, 1.58X) faster as the network becomes more balanced. As a collateral benefit, end-to-end completion times of data-intensive jobs improve as well. Sinbad does so with little impact on the long-term storage balance.
