CuSP

2021 ◽  
Vol 55 (1) ◽  
pp. 47-60
Author(s):  
Loc Hoang ◽  
Roshan Dathathri ◽  
Gurbinder Gill ◽  
Keshav Pingali

Graph analytics systems must analyze graphs with billions of vertices and edges which require several terabytes of storage. Distributed-memory clusters are often used for analyzing such large graphs since the main memory of a single machine is usually restricted to a few hundreds of gigabytes. This requires partitioning the graph among the machines in the cluster. Existing graph analytics systems use a built-in partitioner that incorporates a particular partitioning policy, but the best policy is dependent on the algorithm, input graph, and platform. Therefore, built-in partitioners are not sufficiently flexible. Stand-alone graph partitioners are available, but they too implement only a few policies. CuSP is a fast streaming edge partitioning framework which permits users to specify the desired partitioning policy at a high level of abstraction and quickly generates highquality graph partitions. For example, it can partition wdc12, the largest publicly available web-crawl graph with 4 billion vertices and 129 billion edges, in under 2 minutes for clusters with 128 machines. Our experiments show that it can produce quality partitions 6× faster on average than the state-of-theart stand-alone partitioner in the literature while supporting a wider range of partitioning policies.

2021 ◽  
Vol 5 (ICFP) ◽  
pp. 1-32
Author(s):  
Farzin Houshmand ◽  
Mohsen Lesani ◽  
Keval Vora

Graph analytics elicits insights from large graphs to inform critical decisions for business, safety and security. Several large-scale graph processing frameworks feature efficient runtime systems; however, they often provide programming models that are low-level and subtly different from each other. Therefore, end users can find implementation and specially optimization of graph analytics error-prone and time-consuming. This paper regards the abstract interface of the graph processing frameworks as the instruction set for graph analytics, and presents Grafs, a high-level declarative specification language for graph analytics and a synthesizer that automatically generates efficient code for five high-performance graph processing frameworks. It features novel semantics-preserving fusion transformations that optimize the specifications and reduce them to three primitives: reduction over paths, mapping over vertices and reduction over vertices. Reductions over paths are commonly calculated based on push or pull models that iteratively apply kernel functions at the vertices. This paper presents conditions, parametric in terms of the kernel functions, for the correctness and termination of the iterative models, and uses these conditions as specifications to automatically synthesize the kernel functions. Experimental results show that the generated code matches or outperforms handwritten code, and that fusion accelerates execution.


Author(s):  
Дмитрий Рубвальтер ◽  
Dmitry Rubvalter ◽  
Александр Либкинд ◽  
Alexander Libkind ◽  
Валентина Маркусова ◽  
...  

A multidimensional analysis of the state of Russian studies on the education issues over 1993–2016 was carried out based on the materials of the data contained in the Web of Science (SSCI, A & HCI and SCI-E databases). There were determined the dynamics and trends of a number of relevant indicators, such as the number of Russian publications by year, the share of these publications in the global flow of publications on education issues, the dynamics of the share of publications made in co-authorship with foreign colleagues, etc. A number of distributions of Russian publications on educational issues was compiled and analyzed: by journals, by Russian regions and cities, by organizations and authors of the publications. It was found that most of these distributions were characterized by a high level of non-uniformity. A list of journals (125 titles) in which Russian works on education issues had been published was compiled. Russian organizations (308) and domestic researchers (about two thousand) engaged in studying the issues of education were identified. It was discovered that more than 200 organizations and about 400 academicians from 60 foreign countries had participated in Russian studies on the education issues.


2021 ◽  
Vol 15 (6) ◽  
pp. 1-27
Author(s):  
Marco Bressan ◽  
Stefano Leucci ◽  
Alessandro Panconesi

We address the problem of computing the distribution of induced connected subgraphs, aka graphlets or motifs , in large graphs. The current state-of-the-art algorithms estimate the motif counts via uniform sampling by leveraging the color coding technique by Alon, Yuster, and Zwick. In this work, we extend the applicability of this approach by introducing a set of algorithmic optimizations and techniques that reduce the running time and space usage of color coding and improve the accuracy of the counts. To this end, we first show how to optimize color coding to efficiently build a compact table of a representative subsample of all graphlets in the input graph. For 8-node motifs, we can build such a table in one hour for a graph with 65M nodes and 1.8B edges, which is times larger than the state of the art. We then introduce a novel adaptive sampling scheme that breaks the “additive error barrier” of uniform sampling, guaranteeing multiplicative approximations instead of just additive ones. This allows us to count not only the most frequent motifs, but also extremely rare ones. For instance, on one graph we accurately count nearly 10.000 distinct 8-node motifs whose relative frequency is so small that uniform sampling would literally take centuries to find them. Our results show that color coding is still the most promising approach to scalable motif counting.


2013 ◽  
Vol 20 (1) ◽  
pp. 99-122 ◽  
Author(s):  
Jonathan Fennell

High rates of desertion and surrender during the battles in North Africa in the summer of 1942 were a major factor in Eighth Army’s poor combat performance. At the time, some suggested that these problems were symptomatic of a lack of courage or even of cowardice. There are two broad strands to the conceptualization of courage and cowardice. One focuses on the willingness of the person to fight; the other puts emphasis on how actions express an individual’s ability to cope with fear. Whichever conceptualization is used, high morale motivates the soldier to fight and shields the ordinary recruit from his fear, preventing it from overcoming him in battle. Where morale fails, the soldier is left demotivated and burdened with his terror and, therefore, and is therefore prone to desertion or surrender. Because it is extremely difficult to maintain morale at a continuously high level in an environment governed by chance and managed by humans, all soldiers can find themselves in situations where their actions may be judged as cowardly. Alternatively, if they are properly motivated to fight and prepared by the state and military to deal with the unavoidable fear of combat, all soldiers can be labelled courageous. Accordingly, emotive terms should be avoided when attempting to describe rationally explainable outcomes. The undoubtedly negative connotations attached to cowardice in battle and the positive ones attached to courage are, therefore, arguably unhelpful in understanding Eighth Army’s performance in the summer of 1942 and the human dimension in warfare more generally.


2021 ◽  
Vol 72 (5) ◽  
pp. 110-118
Author(s):  
M. Shkurat ◽  
K. Pavlotska

Analysis of the migration process of the Ukrainian population, which takes place within the state and at the international level during 2014–2019 is carried out in this paper. On the basis of scientific works of domestic scientists and institutes the problems which have not been investigated, and also aspects which have been solved partially are defined. The negative consequences of the Ukrainian migration process on the state of the country are identified. The classification of migration flows of the Ukrainian population according to the reasons of movement is carried out. The main regions and countries which, as of the beginning of 2021, are more attractive to compatriots and are characterized by significant influxes of Ukrainians are identified. The main centers of departure – regions and countries – which are characterized by high level of outflow of Ukrainians are determined, the main problems and reasons that motivate the outflow of Ukrainians are highlighted in this paper. The level of growth of the Ukrainian population in terms of migration flows is defined. The investigation of Ukrainian diasporas, territorial location and analysis of their share in comparison is carried out. The main factors that force the Ukrainian population to move, which affects the demographic situation in Ukraine, the level of employment, as well as the amount of cash flows to the country and the state of the economy as a whole are identified. The main regulations and identified solutions to the migration flows of the domestic population, which were approved and implemented in the period from 2001 to 2019 are compared in this paper. On the basis of current and newly introduced strategies, critical analysis of the ways of regulating the migration process of the Ukrainian population is carried out, the main aspects and problems of the implemented migration policy of Ukraine are determined. The results of the work are to identify the main ways to reduce the negative impact of the migration process on the country's economy and related processes, and strategies to benefit from the movement of citizens of Ukraine, namely: introduction of intellectual security, revision of wages by region, stimulating the process of attracting technology in all spheres of activity and formation of the social protection system.


2002 ◽  
Vol 23 (2) ◽  
pp. 37-55
Author(s):  
Louise I. Shelley

The murder of Valentin Tsvetkov, the governor of Magadan in central Moscow in broad day light in October 2002 highlights that organized crime and corruption are still alive and well and highly destructive of life and governance in Russia (Wines, 2002). His murder once again raises the question, “Why has Russia not been able to stop organized crime and high level corruption?” The answer is that Russia docs not have the political will at the national, regional or local level to fight these problems. This is true because the Kremlin and economic elite push their personal interests over those of the state and the society. Structural problems such as low salaries of state personnel and the embedding of organized crime and corruption make reform very difficult.


1992 ◽  
Vol 114 (4) ◽  
pp. 763-767 ◽  
Author(s):  
J. W. Watts ◽  
T. E. Dwan ◽  
C. G. Brockus

An analog fuel control for a gas turbine engine was compared with several state-space derived fuel controls. A single-spool, simple cycle gas turbine engine was modeled using ACSL (high level simulation language based on FORTRAN). The model included an analog fuel control representative of existing commercial fuel controls. The ACSL model was stripped of nonessential states to produce an eight-state linear state-space model of the engine. The A, B, and C matrices, derived from rated operating conditions, were used to obtain feedback control gains by the following methods: (1) state feedback; (2) LQR theory; (3) Bellman method; and (4) polygonal search. An off-load transient followed by an on-load transient was run for each of these fuel controls. The transient curves obtained were used to compare the state-space fuel controls with the analog fuel control. The state-space fuel controls did better than the analog control.


2018 ◽  
Author(s):  
Ernesto F. L. Amaral

Title in Portuguese: Funções de migração por idade e caracterização de migrantes das microrregiões de Goiás e Distrito Federal, 1975-1979 e 1986-1990(Dissertação de Mestrado)(M.A. Thesis)Since the 70’s, the Brazilian Middle-West Region has experienced an important modernization process in agricultural and industrial activities, which has led to an employment retraction and migration flows, especially from rural areas to the major metropolitan cities of the region. In this research the focus of analysis are the State of Goiás and the Federal District, which encompasses Brasília, the capital of Brazil. In order to understand the new population configuration, the region was divided into four sub-areas (micro-region of Goiânia, micro-region of Entorno de Brasília, the Federal District, and a group of 16 micro-regions of Goiás) and the patterns of migration, by age and sex, were estimated for the periods 1975-1979 and 1986-1991. In addition, differences between native and migrant population of each sub-area, according to the participation in the labor force, income, education and type of migration were investigated. The analysis was performed on the basis of the 1980 and 1991 Brazilian Census data.In the intra-state migration, the fluxes between the micro-region of Goiânia and the 16 other micro-regions of Goiás, and between the Federal District and the micro-region of Entorno de Brasília were the most important. In the inter-state migration, the flux from the states of Mato Grosso, Mato Grosso do Sul and the North region to the 16 micro-regions of Goiás, as well as the flux from Northeast, Southeast and South to the Federal District were significant. But there was a decline in the migration to the Federal District, between 1975-1979 and 1986-1991, while, in the same period, the migration to the Entorno of Brasília increased.The chance of the migrant population in the micro-region of Goiânia and in the Federal District to have worked in the last 12 months prior to census interview was lower than that of the native population. However, those who migrated from the Federal District to the micro-region of Entorno de Brasília had higher chance to have worked in the formal sector than those who did not migrate. Migrants who moved from the micro-region of Entorno de Brasília and the 16 other micro-regions of Goiás to the micro-region of Goiânia had lower wages than the non-migrant population. The lowest level of wages was recorded for the migrants to the 16 other micro-regions of Goiás. Migrants living in the Federal District had high level of education, while those migrants living in the micro-region of Entorno de Brasília had low level of education.In 1986-1990, the micro-regions of Goiânia and Entorno de Brasília, as well as the 16 other micro-regions of Goiás, presented a high percentage of returned migrants. In the same period, a major proportion of migrants to the micro-region of Entorno de Brasília had first moved from Southeast or from the Federal District to others regions. In addition, there was a considerable proportion of migrants with low income who had moved from Northeast to the Federal District and then to the micro-region of the Entorno de Brasília.Future research could focus other areas or regions of Brazil, and could also explore the same kind of analysis, on the basis of the incoming 2000 Brazilian Census, which requires the development of new techniques, as it did not collect information on the city of previous residence, but only on the state of residence.


AGROFOR ◽  
2016 ◽  
Vol 1 (3) ◽  
Author(s):  
Veronika KUKUČKOVÁ ◽  
Nina MORAVČÍKOVÁ ◽  
Radovan KASARDA

The aim of this study was to assess genetic structure of Slovak Pinzgau populationbased on polymorphism at molecular markers using statistical methods. Femaleoffspring of 12 most frequently used bulls in Slovak Pinzgau breeding programmewere investigated. Pinzgau cattle were found to have a high level of diversity,supported by the number of alleles observed across loci (average 5.31, range 2-11)and by the high within-breed expected heterozygosity (average 0.66, range 0.64-0.73). The state of genetic diversity is satisfying and standard for local populations.Detection of 12 possible subpopulation structures provided us with detailedinformation of the genetic structure. The Bayesian approach was applied, detectingthree, as the most probable number of clusters. The similarity of eachsubpopulation using microsatellites was confirmed also by high-throughputmolecular data. The observed inbreeding (FROH=2.3%) was higher than thatexpected based on pedigree data (FPED=0.4%) due to the limited number ofavailable generations in pedigree data. One of the most important steps indevelopment of efficient autochthonous breed protection programs ischaracterization of genetic variability and assessment of the population structure.The chosen set of microsatellites confirmed the suitability in determination of thesubpopulations of Pinzgau cattle in Slovakia. The state of genetic diversity at moredetailed level was successfully performed using bovineSNP50 BeadChip.


2018 ◽  
Vol 24 (1) ◽  
pp. 100
Author(s):  
Lenka Kissiková ◽  
Ivan Dlugoš

<p>The article evaluates the issue of work at heights in industry and reports statistics on fatal accidents at work, the source of which is a fall from above. It also deals with the assessment of the state of personal protective equipment already in use - for example, safety and working ropes and other accessories contaminated with facade paints, lyes, acids or mineral oils and their misuse and dangerous use. The state of the assessed personal protective equipment used was assessed in a test facility on test machines, where the safety of these devices was verified under certain conditions. The article also mentions the issue of inadequate training and training of high-level workers and the lack of training centers that carry out such training.</p>


Sign in / Sign up

Export Citation Format

Share Document