Event Log Mining Tool for Large Scale HPC Systems

Event logging and log files are playing an important role in system and network security. Log files record computer system activities, are used to provide requirements of reliability, security and accountability applications. Information stored in log files can be obtained from different devices, not necessarily clock synchronized, and they do not arrive in the same order they are generated. Nevertheless, log information has to be coherent in time to be useful. To support the events we propose to use Lamport's logic clocks, originated at different sources, in a causal relationship. As a result the administrator will count all the events involved general idea in a computer incident. A model implementation is also presented.

Process mining is based on event logs. Traditionally, an event log is a sequence of events. Yet, there is a growing amount of work in the literature that suggests we should extend the notion of an event log and use partially ordered logs as a basis for process mining. Thus, the need for algorithms able to handle these partially ordered logs will grow in the upcoming years. In this paper, we adapt an existing, classical process discovery algorithm to be able to handle partially ordered logs. We use the ILP Miner [1] as a basis and replace its region theory part by compact tokenflow regions [2] to introduce the ILP2 Miner. This ILP2 Miner handles sequential event logs just like the ILP Miner but, in addition, is able to directly process partially ordered logs. We prove that the ILP2 Miner provides the same guarantees regarding structural and behavioral properties of the discovered process models as the ILP Miner. We implement the ILP2 Miner and show experimental results of its runtime using three well-known example log files from the process mining community literature.

African buffalo optimized multinomial softmax regression based convolutional deep neural network for software fault prediction

One of the most common datasets used by many corporations to gain business intelligence is event log files. Oftentimes, the records in event log files are temporally ordered, and need to be grouped by user ID with the temporal ordering preserved to facilitate mining user behaviors. This kind of analytical workload, here referred to as Relative Order-preserving based Grouping (RE-ORG), is quite common in big data analytics. Using MapReduce/Hadoop for executing RE-ORG tasks on ordered datasets is not efficient due to its internal sort-merge mechanism. In this paper, we propose a distributed framework that adopts an efficient group-order-merge mechanism to provide faster execution of RE-ORG tasks. We demonstrate the advantage of our framework by comparing its performance with Hadoop through extensive experiments on real-world datasets. The evaluation results show that our framework can achieve up to 6.3x speedup over Hadoop in executing RE-ORG tasks.

In large scale systems IT systems, for the purpose of management, we have what are known as Networked Operations Centers or NOCs. These NOCs are manned by support staff, known as operators. Data in the form of events from various sub-systems in the larger system is sent to the central event console so that the NOC operators can work on problems in the infrastructure before they have a significant impact on the availability and performance of business applications. The challenge for these NOC operators is that of managing the large number of incoming events. In this work, we address the problem of scale by proposing a data mining technique for automatic event correlation for large scale operations management systems.

Numerical estimation has been used to study how children mentally represent numbers for many years (e.g., Siegler & Opfer, 2003). However, these studies have always presented children with positive numbers and positive number lines. Children’s mental representation of negative numbers has never been addressed. The present study tested children in the 2nd, 4th, and 6th grades to assess their mental representations of both positive and negative numbers using a standard numerical estimation task. We replicated the shift from a logarithmic to linear representation for positive numbers (0–1,000 scale) in that 2nd graders represented positive numbers logarithmically, but 4th and 6th graders represented the numbers linearly. Furthermore, children’s representation of negative numbers paralleled their representations of positive numbers and showed the same shift from a logarithmic representation at Grade 2 to linear representations at Grades 4 and 6. This is the first study to provide data on children’s representation of negative numbers, and the implications of these findings are discussed.

In this paper, we present a variance minimization (VM) procedure for rare event simulation in tandem queueing networks. We prove that the VM method can produce a zero variance. The VM method is suitable to compute optimal importance sampling (IS) parameters for small scale tandem networks. For large scale tandem networks we propose a sub-optimal IS (SOIS) method, which projects the optimal biased transition probabilities of the corresponding small scale system into those of a large scale system. In other words, we establish an efficient IS method for a large scale system by zooming into a small scale system and then projecting our findings into the large scale system. The numerical results show that our SOIS method can produce accurate results with very short CPU time, while many other methods often require much longer.

Log files contain valuable information about the execution of a system. This information is often used for debugging, operational profiling, finding anomalies, detecting security threats, measuring performance etc. The log files are usually too big for extracting this valuable information manually, even though manual perusal is still one of the more widely used techniques. Recently a variety of data mining and machine learning algorithms are being used to analyze the information in the log files. A major road block for the efficient use of these algorithms is the inherent variability present in every log line of a log file. Each log line is a combination of a static message type field and a variable parameter field. Even though both these fields are required, the analyses algorithm often requires that these be separated out, in order to find correlations in the repeating log event types. This disentangling of the message and parameter fields to find the event types is called abstraction of log lines. Each log line is abstracted to a unique ID or event type and the dynamic parameter value is extracted to give an insight on the current state of the system. In this paper we present a technique based on a clustering technique used in the Simple Log file Clustering Tool for log file abstraction. This solution is especially useful when we don't have access to the source code of the application or when the lines in the log file do not conform to a rigid structure. We evaluated our implementation on log files from the Virtual Computing Lab, a cloud computer management system at North Carolina State University, and abstracted it to 727 unique event types.

4 - POSITIVES AND NEGATIVES

The implementation of large-scale renewable energy systems (such as solar and wind) in the GCC is widely growing nowadays, particularly in remote areas generation applications where the grid extension is costly. The integration of renewable energy systems into the existing power grids, especially the hybrid ones like hybrid solar wind systems, not only provide a clean environmentally friendly supply of electricity but also provide a more reliable supply of electricity through the utilization of different energy sources. However, the operating characteristics of such renewable energy systems are strongly influenced by the weather patterns, i.e. intermittent generation, and will therefore impact the operation of the power grid into which they are integrated. Certain amounts of renewable distributed generation can be integrated into the existing grid without affecting its operation; however, the massive introduction of large-scale renewable energy systems into the grid will result into an increased level of risk of overload and over voltage, unacceptable levels of power quality and reliability, together with the possibility of incorrect operation of the protection system. Therefore, significant investments will be needed to enable the existing power grids to accommodate such a large scale of renewable energy systems into it while avoiding such impacts.In this paper, a comprehensive study of the impact of integrating large-scale hybrid renewable energy systems into Qatar's electrical power grid is carried out. The study includes investigating the grid power flow, power quality, efficiency, transient stability, and reliability under both normal and contingency operating conditions. The study aims to give solutions for allowing the increase in large scale hybrid renewable energy systems penetration into Qatar's grid while reducing their influence and maintaining the grid reliability and stability during all operating conditions. The study is performed based on Qatar's grid real-time operating conditions, and takes into account Qatar's climate profile (e.g. Qatar average annual wind speed and solar irradiation) in order to simulate the intermittent output of the integrated hybrid renewable energy system. Moreover, an economic study and an environmental impact study are carried out to evaluate the large-scale integration of hybrid renewable energy systems into Qatar's grid from different aspects. In summary, the carried out studies will show the impact of implementing large-scale hybrid renewable energy systems on the performance of Qatar's power grid, together with its environmental and economic benefits. The real-time implementation will suggest how to introduce the large scale hybrid renewable energy systems gradually into Qatar's grid, and will propose the best scenario for operating both the existing conventional stations and the integrated hybrid systems while balancing their electricity generation with the demand. The real-time implementation utilizes a new smart control and management mechanism that is proposed to allow such a balance between the generation and demand on introducing large-scale hybrid renewable energy systems into Qatar's existing power grid. The outcome of this study will help Qatar in assessing and evaluating the operational, environmental and economic aspects of large scale hybrid renewable energy systems penetration into Qatar's power grid, and will demonstrate the challenges that will face Qatar when integrating such a large capacity into its grid.

The COVID-19 pandemic, acceleration of climate change, and geopolitical turmoil have triggered discussions about how public health systems in Europe need to transform to be adequately equipped for the so-called “permacrisis”. The challenges to policy makers, health institutions and health services have become so multifaceted, that single solutions from the health sector alone no longer work. Yet, despite the complexity, countries seek to get its own house in order whilst building on inter-sectoral and European approaches in parallel. Germany, Ireland, and Ukraine have all embarked upon ambitious public health services and institutional reforms, leading to transformative change in public health systems. WHO has been accompanying these reforms. The workshop will provide a platform to draw insights and lessons on transforming public health systems from all three countries, although their population size, their socioeconomic and political situation and their public health system design is so different. As the smallest of the three countries in population, Ireland created coherence and is aligning public health actors within “Healthy Ireland” and the Sláintecaire and the Crowe Horwath public health reforms. Germany’s “Pact for the Public Health Services” has been one of the greatest public health investment programs in Europe that was initiated following lessons from the COVID-19 pandemic. And Ukraine has undertaken important legal steps to adopt a Public Health System Law, in the middle of a war of aggression on its territory. The workshop will exhibit the process of large-scale public health system transformation in each of the three countries. The purpose is to identify the enablers that make large scale system transformation possible and addressing some of the pains and gains. The workshop will have three parts. Part one consists of three presentations on large scale system reforms from each of the participating countries. Part two consists of a moderated Q&A session. Part three will focus on the synthesis of enablers for large scale public health system reforms through interaction with the workshop participants, using a digital white board. Key messages • Disruptive times can create enabling conditions for large scale public health system transformation. • Designing and implementing large scale public health system reforms is complex, requiring passion, resilience, and collaborative leadership.

The emergence of large scale systems provides solutions and poses new challenges to a number of existing problems covering a broad spectrum of applications. Among these problems, we are interested by exploring new approaches to integrate data in large scale systems. For this purpose, we propose a data integration framework that is data-structure-independent, allowing a high level of semantic interoperability. We describe the methods used in our framework to build an index regrouping all the clusters published data descriptions that are designed to fit the structure of a large scale system. This framework focuses on the properties of large scale systems such as multi-level heterogeneity and high degree of dynamicity and evolution.

Fine-Grained Parallel and Distributed Spatial Stochastic Simulation of Biological Reactions

Computer Aided Design of the Stochastic Controller for a Large Scale River System

Identification, Segmentation, and Image Property Study of Acute Infarcts in Diffusion-Weighted Images by Using a Probabilistic Neural Network and Adaptive Gaussian Mixture Model