Towards Analyzing Computational Costs of Spark for SARS-CoV-2 Sequences Comparisons on a Commercial Cloud

In this paper, we developed a Spark application, named Diff Sequences Spark, which compares 540 SARS-CoV-2 sequences from South America in Amazon EC2 Cloud, generating as output the positions where the differences occur. We analyzed the performance of the proposed application on selected memory and storage optimized virtual machines (VMs) at on-demand and spot markets. The execution times and financial costs of the memory optimized VMs outperformed the storage optimized ones. Regarding the markets, Diff Sequences Spark reduced the average execution times and monetary costs when using spot VMs compared to their respective on-demand VMs, even in scenarios with several spot revocations, benefiting from the low overhead fault tolerance Spark framework.

RACIONALIZACIÓN DE LOS PROCESOS EN LA REVISIÓN FÍSICA DE INVENTARIOS EN EL ÁREA DE ALMACENAMIENTO DE PRODUCTOS TERMINADOS GARANTIZANDO LA SEGURIDAD Y SALUD EN EL TRABAJO

The company on which this study was carried out performs a technical review of the inventory once a month, with a level of confidence and accuracy of 100%, although its level of precision is very high, it is necessary to change the how nventory control is being executed, since the current method exposes the operator to unnecessary and infallible risk, which has motivated the operations manager to request a change in the process. However, the process with which the inventory is currently being verified requires more man hours and the payment of overtime to its workers to complete the activity, generating an increase in nominal costs. For this reason, the objective of this work is to optimize the method of operation in the physical review of the inventory, allowing it to improve its effectiveness, reduce nominal costs, execution times and guaranteeing safety and health at work. Therefore, the work was carried out through descriptive applied research, developing strategies that start from theories, in order to apply them and thus give an optimal solution to the problems found in the company. Likewise, a quantitative investigation was implemented where the data emerged from a field work carried out in the finished product storage area, through a study of times and movements in the inventory review. Additionally, a cost analysis was carried out, accompanied by various simulations that made it possible to establish which is the appropriate machinery for the process. As a result, it was determined that the drone and the zigzag method is the optimal solution for the physical review of inventory by reducing risks, costs and execution times.

Benchmarking geospatial database on Kubernetes cluster

Kubernetes is an open-source container orchestration system for automating container application operations and has been considered to deploy various kinds of container workloads. Traditional geo-databases face frequent scalability issues while dealing with dense and complex spatial data. Despite plenty of research work in the comparison of relational and NoSQL databases in handling geospatial data, there is a shortage of existing knowledge about the performance of geo-database in a clustered environment like Kubernetes. This paper presents benchmarking of PostgreSQL/PostGIS geospatial databases operating on a clustered environment against non-clustered environments. The benchmarking process considers the average execution times of geospatial structured query language (SQL) queries on multiple hardware configurations to compare the environments based on handling computationally expensive queries involving SQL operations and PostGIS functions. The geospatial queries operate on data imported from OpenStreetMap into PostgreSQL/PostGIS. The clustered environment powered by Kubernetes demonstrated promising improvements in the average execution times of computationally expensive geospatial SQL queries on all considered hardware configurations compared to their average execution times in non-clustered environments.

Double auction and profit maximization mechanism for jobs with heterogeneous durations in cloud federations

By sharing resources with each other, different cloud providers in a cloud federation can exploit their diversity in resource configuration and operational cost so as to improve service performance. They should consider the strategy of resource pricing, job scheduling and server provisioning altogether to maximize their own interests. On the other hand, they need to efficiently trade the resources with a suitable mechanism, typically auction, so as to guarantee the participants’ profits. Nevertheless, in consideration of the heterogeneous execution times of jobs, both the pricing strategy and trading mechanism should be delicately designed, which is obviously a challenging task. In this paper, we firstly propose a truthful, individual-rational and ex-post budget-balanced auction mechanism for selecting pairs of buyer and seller winners to trade virtual machines for different durations. Then, to maximize the individual profits, we propose a dynamic resource bidding scheme and a job scheduling strategy based on our importance model of jobs with heterogeneous execution times and resource requirements. The simulation results show that, compared with existing ones, our design can better handle varieties of both execution time and resource requirement and make the participants obtain more individual profits.

An Approach to the Analysis of Causes of Delays in Industrial Construction Projects through Planning and Statistical Computing

The analysis of the planning activities of industrial construction projects can help to evaluate some of the causes that have an impact on the variation of execution times and can also contribute to identifying those activities and components that are most likely to experience or cause delays. Data analysis is facilitated by the use of techniques based on statistical programs, allowing delays to be unequivocally linked to the different elements that make up these projects. In a theoretical study, a simulation is carried out with data that are hypothetical but consistent with real projects, which are transformed and standardized before being uploaded to the statistical software. Using the statistical software’s graphical interface, the data set is analyzed from a descriptive point of view, unraveling the relationships between variables and factors by means of contingency tables and scatter plots. Using other techniques such as the comparison of variables and correlation studies, as well as linear regression and variance analysis, the characteristics are evaluated and the differences in project delays are investigated in order to determine, after the fact, which components have the highest rates of delay in execution times.

Image Compression Using Permanent Neural Networks for Predicting Compact Discrete Cosine Transform Coefficients

This study proposes a new image compression technique that produces a high compression ratio yet consumes low execution times. Since many of the current image compression algorithms consume high execution times, this technique speeds up the execution time of image compression. The technique is based on permanent neural networks to predict the discrete cosine transform partial coefficients. This can eliminate the need to generate the discrete cosine transformation every time an image is compressed. A compression ratio of 94% is achieved while the average decompressed image peak signal to noise ratio and structure similarity image measure are 22.25 and 0.65 respectively. The compression time can be neglected when compared to other reported techniques because the only needed process in the compression stage is to use the generated neural network model to predict the few discrete cosine transform coefficients.