Benchmarking of distributed computing engines spark and GraphLab for big data analytics

In this paper we evaluate and compare two representativeand popular distributed processing engines for large scalebig data analytics, Spark and graph based engine GraphLab. Wedesign a benchmark suite including representative algorithmsand datasets to compare the performances of the computingengines, from performance aspects of running time, memory andCPU usage, network and I/O overhead. The benchmark suite istested on both local computer cluster and virtual machines oncloud. By varying the number of computers and memory weexamine the scalability of the computing engines with increasingcomputing resources (such as CPU and memory). We also runcross-evaluation of generic and graph based analytic algorithmsover graph processing and generic platforms to identify thepotential performance degradation if only one processing engineis available. It is observed that both computing engines showgood scalability with increase of computing resources. WhileGraphLab largely outperforms Spark for graph algorithms, ithas close running time performance as Spark for non-graphalgorithms. Additionally the running time with Spark for graphalgorithms over cloud virtual machines is observed to increaseby almost 100% compared to over local computer clusters.

Perturbative discrete-time multivariate fiber channel model with finite memory

We introduce a discrete-time fibre channel model that provides an accurate analytical description of signal-signal and signal-noise interference with memory defined by the interplay of nonlinearity and dispersion. Also the conditional pdf of signal distortion, which captures non-circular complex multivariate symbol interactions, is derived providing the necessary platform for the analysis of channel statistics and capacity estimations in fibre optic links.

Metabolic memory effect of the saturated fatty acid, palmitate, in monocytes

Hyperglycaemia has a deferred detrimental effect on glucose metabolism, termed "metabolic memory". Elevated saturated fatty acids promote insulin resistance, hyperglycaemia and associated atherosclerotic complications, but their effect on "metabolic memory" is unknown. Therefore we investigated whether basal and insulin-stimulated (10(-6)M for 12h) glucose (2-deoxy-D-[(3)H]-glucose) uptake was affected by palmitate pre-treatment human THP-1 monocytes. Palmitate-induced a time-dependent and concentration-dependent inhibition of insulin-stimulated glucose uptake, showing almost complete abolition of the insulin-stimulatory effect with 300 microM palmitate. Basal glucose uptake was unaffected by palmitate. When palmitate was washed out, the inhibitory effect on insulin-stimulated glucose uptake persisted for at least 60 h.

Three-factor models versus time series models:quantifying time-dependencies of interactions between stimuli in cell biology and psychobiology for short longitudinal data

Signal integration determines cell fate on the cellular level, affects cognitive processes and affective responses on the behavioural level, and is likely to be involved in psychoneurobiological processes underlying mood disorders. Interactions between stimuli may subjected to time effects. Time-dependencies of interactions between stimuli typically lead to complex cell responses and complex responses on the behavioural level. We show that both three-factor models and time series models can be used to uncover such time-dependencies. However, we argue that for short longitudinal data the three factor modelling approach is more suitable. In order to illustrate both approaches, we re-analysed previously published short longitudinal data sets. We found that in human embryonic kidney 293 cells cells the interaction effect in the regulation of extracellular signal-regulated kinase (ERK) 1 signalling activation by insulin and epidermal growth factor is subjected to a time effect and dramatically decays at peak values of ERK activation. In contrast, we found that the interaction effect induced by hypoxia and tumour necrosis factor-alpha for the transcriptional activity of the human cyclo-oxygenase-2 promoter in HEK293 cells is time invariant at least in the first 12-h time window after stimulation. Furthermore, we applied the three-factor model to previously reported animal studies. In these studies, memory storage was found to be subjected to an interaction effect of the beta-adrenoceptor agonist clenbuterol and certain antagonists acting on the alpha-1-adrenoceptor / glucocorticoid-receptor system. Our model-based analysis suggests that only if the antagonist drug is administer in a critical time window, then the interaction effect is relevant.

Memory effects in a non-equilibrium growth model

We study memory effects in a kinetic roughening model. For d=1, a different dynamic scaling is uncovered in the memory dominated phases; the Kardar-Parisi-Zhang scaling is restored in the absence of noise. dc=2 represents the critical dimension where memory is shown to smoothen the roughening front (a=0). Studies on a discrete atomistic model in the same universality class reconfirm the analytical results in the large time limit, while a different scaling behavior shows up for t

A working memory bias for alcohol-related stimuli depends on drinking score

We tested 44 participants with respect to their working memory (WM) performance on alcohol-related versus neutral visual stimuli. Previously an alcohol attentional bias (AAB) had been reported using these stimuli, where the attention of frequent drinkers was automatically drawn toward alcohol-related items (e.g., beer bottle). The present study set out to provide evidence for an alcohol memory bias (AMB) that would persist over longer time-scales than the AAB. The WM task we used required memorizing 4 stimuli in their correct locations and a visual interference task was administered during a 4-sec delay interval. A subsequent probe required participants to indicate whether a stimulus was shown in the correct or incorrect location. For each participant we calculated a drinking score based on 3 items derived from the Alcohol Use Questionnaire, and we observed that higher scorers better remembered alcohol-related images compared with lower scorers, particularly when these were presented in their correct locations upon recall. This provides first evidence for an AMB. It is important to highlight that this effect persisted over a 4-sec delay period including a visual interference task that erased iconic memories and diverted attention away from the encoded items, thus the AMB cannot be reduced to the previously reported AAB. Our finding calls for further investigation of alcohol-related cognitive biases in WM, and we propose a preliminary model that may guide future research. © 2012 American Psychological Association.

Long memory and multifractality:a joint test

The properties of statistical tests for hypotheses concerning the parameters of the multifractal model of asset returns (MMAR) are investigated, using Monte Carlo techniques. We show that, in the presence of multifractality, conventional tests of long memory tend to over-reject the null hypothesis of no long memory. Our test addresses this issue by jointly estimating long memory and multifractality. The estimation and test procedures are applied to exchange rate data for 12 currencies. Among the nested model specifications that are investigated, in 11 out of 12 cases, daily returns are most appropriately characterized by a variant of the MMAR that applies a multifractal time-deformation process to NIID returns. There is no evidence of long memory.

What checkers actually check:an eye tracking study of inhibitory control and working memory

Background - Not only is compulsive checking the most common symptom in Obsessive Compulsive Disorder (OCD) with an estimated prevalence of 50–80% in patients, but approximately ~15% of the general population reveal subclinical checking tendencies that impact negatively on their performance in daily activities. Therefore, it is critical to understand how checking affects attention and memory in clinical as well as subclinical checkers. Eye fixations are commonly used as indicators for the distribution of attention but research in OCD has revealed mixed results at best. Methodology/Principal Finding - Here we report atypical eye movement patterns in subclinical checkers during an ecologically valid working memory (WM) manipulation. Our key manipulation was to present an intermediate probe during the delay period of the memory task, explicitly asking for the location of a letter, which, however, had not been part of the encoding set (i.e., misleading participants). Using eye movement measures we now provide evidence that high checkers’ inhibitory impairments for misleading information results in them checking the contents of WM in an atypical manner. Checkers fixate more often and for longer when misleading information is presented than non-checkers. Specifically, checkers spend more time checking stimulus locations as well as locations that had actually been empty during encoding. Conclusions/Significance - We conclude that these atypical eye movement patterns directly reflect internal checking of memory contents and we discuss the implications of our findings for the interpretation of behavioural and neuropsychological data. In addition our results highlight the importance of ecologically valid methodology for revealing the impact of detrimental attention and memory checking on eye movement patterns.

An improved memory management scheme for large scale graph computing engine GraphChi

GraphChi is the first reported disk-based graph engine that can handle billion-scale graphs on a single PC efficiently. GraphChi is able to execute several advanced data mining, graph mining and machine learning algorithms on very large graphs. With the novel technique of parallel sliding windows (PSW) to load subgraph from disk to memory for vertices and edges updating, it can achieve data processing performance close to and even better than those of mainstream distributed graph engines. GraphChi mentioned that its memory is not effectively utilized with large dataset, which leads to suboptimal computation performances. In this paper we are motivated by the concepts of 'pin ' from TurboGraph and 'ghost' from GraphLab to propose a new memory utilization mode for GraphChi, which is called Part-in-memory mode, to improve the GraphChi algorithm performance. The main idea is to pin a fixed part of data inside the memory during the whole computing process. Part-in-memory mode is successfully implemented with only about 40 additional lines of code to the original GraphChi engine. Extensive experiments are performed with large real datasets (including Twitter graph with 1.4 billion edges). The preliminary results show that Part-in-memory mode memory management approach effectively reduces the GraphChi running time by up to 60% in PageRank algorithm. Interestingly it is found that a larger portion of data pinned in memory does not always lead to better performance in the case that the whole dataset cannot be fitted in memory. There exists an optimal portion of data which should be kept in the memory to achieve the best computational performance.

On the k-anonymization of time-varying and multi-layer social graphs

The popularity of online social media platforms provides an unprecedented opportunity to study real-world complex networks of interactions. However, releasing this data to researchers and the public comes at the cost of potentially exposing private and sensitive user information. It has been shown that a naive anonymization of a network by removing the identity of the nodes is not sufficient to preserve users’ privacy. In order to deal with malicious attacks, k -anonymity solutions have been proposed to partially obfuscate topological information that can be used to infer nodes’ identity. In this paper, we study the problem of ensuring k anonymity in time-varying graphs, i.e., graphs with a structure that changes over time, and multi-layer graphs, i.e., graphs with multiple types of links. More specifically, we examine the case in which the attacker has access to the degree of the nodes. The goal is to generate a new graph where, given the degree of a node in each (temporal) layer of the graph, such a node remains indistinguishable from other k-1 nodes in the graph. In order to achieve this, we find the optimal partitioning of the graph nodes such that the cost of anonymizing the degree information within each group is minimum. We show that this reduces to a special case of a Generalized Assignment Problem, and we propose a simple yet effective algorithm to solve it. Finally, we introduce an iterated linear programming approach to enforce the realizability of the anonymized degree sequences. The efficacy of the method is assessed through an extensive set of experiments on synthetic and real-world graphs.

Long memory and multifractality:a joint test

The properties of statistical tests for hypotheses concerning the parameters of the multifractal model of asset returns (MMAR) are investigated, using Monte Carlo techniques. We show that, in the presence of multifractality, conventional tests of long memory tend to over-reject the null hypothesis of no long memory. Our test addresses this issue by jointly estimating long memory and multifractality. The estimation and test procedures are applied to exchange rate data for 12 currencies. In 11 cases, the exchange rate returns are accurately described by compounding a NIID series with a multifractal time-deformation process. There is no evidence of long memory.

Receiver memory requirement in mode delay compensated few-mode fibre spans with intermediate coupling

The required receiver time window after propagation through few-mode fibre is studied for a broad range of coupling and mode delay span configurations. Under intermediate coupling, effective mode delay compensation is observed for a compensation period of 25km.