Server Performance Research Articles

Predictive Modelling For Credit Card Fraud Detection Using Data Analytics

The finance and banking is very important sector in our present day generation, where almost every human has to deal with bank either physically or online [10]. The productivity and profitability of both public and private sector has tremendously increased because of banking information system. Nowadays most of E-commerce application system transactions are done through credit card and online net banking. These systems are vulnerable with new attacks and techniques at alarming rate. Fraud detection in banking is one of the vital aspects nowadays as finance is major sector in our life. As data is increasing in terms of Peta Bytes (PB) and to improve the performance of analytical server in model building, we have interface analytical framework with Hadoop which can read data efficiently and give to analytical server for fraud prediction. In this paper we have discussed a Big data analytical framework to process large volume of data and implemented various machine learning algorithms for fraud detection and observed their performance on benchmark dataset to detect frauds on real time basis there by giving low risk and high customer satisfaction.

Firing Rate Optimization of Deterministic Timed Event Graphs by Server Performance Improvement

Timed event graphs are a subclass of Petri nets that are commonly applied for modeling and controlling manufacturing systems. In this paper, we investigate the problem of improving the performance of machines (servers) in timed event graphs in order to maximize the firing rate of a system. We show that the logic constraints on the cost of machines and the operation time of machines (firing delay of transitions) can be transformed into linear algebraic constraints. Furthermore, we formulate a mixed integer linear programming problem to maximize the firing rate of the net under a given budget for improving the performance of servers (decreasing the firing delay of transitions) that can provide an optimal solution. Finally, application to an assembly line is presented to show the effectiveness of the developed methodology.

On the Performance Evaluation of IaaS Cloud Services With System-Level Benchmarks

Infrastructure-as-a-service is a cloud service model that allows customers to outsource computing resources such as servers and storage. This article evaluates four IaaS cloud services - Amazon EC2, Microsoft Azure, Google Compute Engine and Rackspace Cloud in a vendor-neutral approach with regards to system parameter usage including server, file I/O and network utilization. Thus, system-level benchmarking provides objective comparison of cloud providers from performance standpoint. Unixbench, Dbench and Iperf are the System-level benchmarks chosen to test the performance of server, file I/O and network respectively. In order to capture the variation in performance, the tests were performed at different times on weekdays and weekends. With each offering, the benchmarks are tested on different configurations to provide an insight to the cloud users in selection of provider followed by appropriate VM sizing according to the workload requirement. In addition to the performance evaluation, price-per-performance value of all the providers is also examined and compared.

Analysis and Simulink Modeling of the Performance of Dynamic Web Server Using JSP and PHP

In recent years, web technology carried on growing and at same time, the number of internet users increased significantly in number. Today, a Web server is capable of processing millions of requests per day, but during the peak period may collapse and becomes critical causing unavailability of the services offered by the servers. That is why Web server performance is a topic of great interest to many researchers. In this paper, we evaluate experimentally the impact of JSP and PHP dynamic content technology: JSP and PHP with access to a database of performance data of Apache Web server. Using the “ApacheBench” performance measurement tool, the approach is to compare the performances of four different configurations of a Web server, such as: Apache Web server implementing JSP technology with access to PostgreSQL database, Apache using PHP technology with the PostgreSQL as database, Apache Web server using the JSP technology with access to MySQL database, finally Apache and PHP with DBMS MySQL. At the end of this article, we also present a Simulink model of Web server performance based on the simple M/M/1 queue. During the modeling, the MATLAB software was used.

Database problems of maritime transport industry on high load platform

The article says about distributed and high load systems, common problems and common instructions for solving such problems as denormalization, replication and sharding of transport and maritime aspects. The analysis of strengths and weaknesses of technology for the transport field of activity was carried out. A fragment of the code for sampling optimization is given. Conclusions about the global use of industry technology are made. When your current infrastructure starts to show the first signs that it is no longer cope with the load and if you have a 128 MB VPS - for you, it can be 10 requests per second. For corporations, it can be 10,000 requests. There is a need for scaling and optimizing the infrastructure. Maritime office needs a good monitoring system. It will help determine the very moment when it's time to scale. Stable systems for monitoring and tracking server performance trends: Munin, Zabbix, Nagios. What usually happens when the high load point is approached: slow or infinite loading of pages; random errors; broken connections from the Web server; partial loading of content (for example, there is no part of the image), and a decrease in audience activity. This direction is very modern and promising for the transport industry.

Analisis Perbandingan Kinerja Web Server Apache dan Nginx Menggunakan Httperf Pada Portal Berita (Studi Kasus beritalinux.com)

The internet world in the globalization era is now developing. Anyone and anywhere can access the internet if you already have the tools and connections are adequate. There are two-ways relationship in accessing a web, they are Client and Server. Good Web Server performance can affect the quality of two-ways relationship between Client and Server. There are two Web Servers that are widely used today are Apache and Nginx. As a media content provider is expected to meet all the needs of users, especially in terms of performance of the device itself. To prove how the apache and nginx Web Server performance compare to the data request by the user, it is necessary to do a test and compare the parameters of each Web Server. Based on the problems above, the authors want to analyze and compare the performance of both Web servers are Apache and Nginx, so users can choose the best Web Server. The author only compares the parameters of throughput, connection, request, reply and error by assigning load to each test and performed on attributes that exist on beritalinux.com virtually. After testing, the results obtained where in responding and connecting data that requested by the client of web application server nginx was superior to apache. From these results, it is recommended to admin on beritalinux.com to use nginx web server for better website performance.Keywords: Analysis, Performance, Web Server, Apache, Nginx, HTTPERF

Measurement of Energy Performances for General-Structured Servers

Energy consumption of servers in data centers increases rapidly along with the wide application of Internet and connected devices. To improve the energy efficiency of servers, voluntary or mandatory energy efficiency programs for servers, including voluntary label program or mandatory energy performance standards have been adopted or being prepared in the US, EU and China. However, the energy performance of servers and testing methods of servers are not well defined. This paper presents matrices to measure the energy performances of general-structured servers. The impacts of various components of servers on their energy performances are also analyzed. Based on a set of normalized workload, the author proposes a standard method for testing energy efficiency of servers. Pilot tests are conducted to assess the energy performance testing methods of servers. The findings of the tests are discussed in the paper.

Definition and classification of evaluation units for tertiary structure prediction in CASP12 facilitated through semi-automated metrics.

For assessment purposes, CASP targets are split into evaluation units. We herein present the official definition of CASP12 evaluation units (EUs) and their classification into difficulty categories. Each target can be evaluated as one EU (the whole target) or/and several EUs (separate structural domains or groups of structural domains). The specific scenario for a target split is determined by the domain organization of available templates, the difference in server performance on separate domains versus combination of the domains, and visual inspection. In the end, 71 targets were split into 96 EUs. Classification of the EUs into difficulty categories was done semi-automatically with the assistance of metrics provided by the Prediction Center. These metrics account for sequence and structural similarities of the EUs to potential structural templates from the Protein Data Bank, and for the baseline performance of automated server predictions. The metrics readily separate the 96 EUs into 38 EUs that should be straightforward for template-based modeling (TBM) and 39 that are expected to be hard for homology modeling and are thus left for free modeling (FM). The remaining 19 borderline evaluation units were dubbed FM/TBM, and were inspected case by case. The article also overviews structural and evolutionary features of selected targets relevant to our accompanying article presenting the assessment of FM and FM/TBM predictions, and overviews structural features of the hardest evaluation units from the FM category. We finally suggest improvements for the EU definition and classification procedures.

An efficient scheduling multimedia transcoding method for DASH streaming in cloud environment

As a result of technological evolution, streaming service providers have been dealing with the problem of delivery multimedia content to the diversity of devices with different resolutions. This issue can be solved by using dynamic adaptive streaming over hypertext (DASH) transfer protocol. However, a transcoding job in DASH requires a lot of computation resource which could lead to delaying the starting of multimedia streaming. Recently, new studies have addressed novel scheduling methods on video transcoding, but those research did not solve the problem entirely, such as the solution did not concern server performance or speed connection between a server and its requested users. Moreover, the load and speed connection status of the data servers is often unstable, leading to increasing the starting delay. So in this article, we solve such problem by modeling transcoding jobs in the form of an optimization problem and propose an algorithm to find an optimal schedule to transcode video source files. In which, we use moving average method to find average points for a short period to deal with server state changes. In the experiment, we implement our proposed method with DASH to demonstrate the effectiveness of the optimization scheduling method. In the system, we create several servers running on the Docker platform to simulate a cloud environment. Experimental results show that our methodology reduces the time of the transcoding process up to 30% compared to existing research.

Defensive Mechanism Against DDoS Attack to Preserve Resource Availability for IoT Applications

The major challenge of Internet of Things (IoT) generated data is its hypervisor level vulnerabilities. Malicious VM deployment and termination are so simple due to its multitenant shared nature and distributed elastic cloud features. These features enable the attackers to launch Distributed Denial of Service attacks to degrade cloud server performance. Attack detection techniques are applied to the VMs that are used by malicious tenants to hold the cloud resources by launching DDoS attacks at data center subnets. Traditional dataflow-based attack detection methods rely on the similarities of incoming requests which consist of IP and TCP header information flows. The proposed approach classifies the status patterns of malicious VMs and ideal VMs to identify the attackers. In this article, information theory is used to calculate the entropy value of the malicious virtual machines for detecting attack behaviors. Experimental results prove that the proposed system works well against DDoS attacks in IoT applications.

Improved protein contact predictions with the MetaPSICOV2 server in CASP12.

In this paper, we present the results for the MetaPSICOV2 contact prediction server in the CASP12 community experiment (http://predictioncenter.org). Over the 35 assessed Free Modelling target domains the MetaPSICOV2 server achieved a mean precision of 43.27%, a substantial increase relative to the server's performance in the CASP11 experiment. In the following paper, we discuss improvements to the MetaPSICOV2 server, covering both changes to the neural network and attempts to integrate contact predictions on a domain basis into the prediction pipeline. We also discuss some limitations in the CASP12 assessment which may have overestimated the performance of our method.

RANCANG BANGUN LOAD BALANCING APACHE WEB SERVER DAN DATABASE CLUSTER PADA BAGIAN ADMINISTRASI PEMBANGUNAN DAN DATA ELEKTRONIK KOTA PEKANBARU

ABSTRACT
 The Central Statistics Agency (BPS) in collaboration with the Association of Indonesian Internet Service Provider (APJII) recorded a growth rate of Internet users in Indonesia by the end of 2013 had reached 71.19 million people. The increase of Internet users who access the website very significant every day, with a load demand (request) is varied. This causes the performance of a web server be increased. Load balancing is a technique to distribute the traffic load on two or more lines in a balanced connections, so that traffic can run optimally, maximize throughput, minimize response time and avoid overload on one connection line. Now, database is required in order to run fast, have high reliability and availability, the database stored clustering can be divided into multiple machines and at the time the application is running, all the machines that store the data are considered as a single entity.
 From this research, author can concludes. First, Load balancing concept can give solutions to build a web server that can handle the request is quite high and quality in order to improve the quality of government services. With the web server load balancer service remains available when one server goes down, the other servers will automatically replace it, so the user can continue to access as if there is no problem on the server. Second, Data Base using clustering method is excellent for loadbalancing in the handling of system failure due to the ability of each machine will be used and if there is one machine that experienced a failure, the system will not be directly disturbed because other machines will still work.

Power Efficiency Study Using SPECjEnterprise2010

In this paper, we present a case study of the power consumption and performance tradeoffs in Java application servers. We run SPECjEnterprise2010 on two platforms with different CPUs, namely, AMD Phenom II and Intel Atom. We investigate the performance and power consumption behaviors against the increasing system size and the relative performance between two CPUs. We also study the effectiveness of the dynamic frequency scaling (DFS) on Phenom and HyperThreading (HT) on Atom. Phenom achieves 8.5 times higher transaction throughput with 18.6 times more dynamic power consumption: Thus, the relative power consumption of Phenom against Atom is 218%. DFS parameter tuning reduces the dynamic power consumption of the Phenom-based server by up to 31%. However, it also has a side effect of higher transaction response time when the load level is fluctuated. HT improves the performance of the Atom server by up to 47% (standard workload) or 70% (individual transaction). HT does not reduce the power consumption of standard workload, but it works for individual transactions by up to 30% reduction of normalized power.

Analysis of Queue System to Improve the Quality of Service in Gra PARI Telkomsel Banda Aceh

The problem of queue is how to balance the waiting costs and preventing cost of the queue to gain maximumprofit. Queuing analysis can be done by knowing arrival distribution, service time distribution, queue discipline, andqueuing model. The purpose of the paper is to calculate the optimal number of servers so users do not wait too longin line, know the queue model in accordance with queuing conditions, and improve the efficiency and performance ofservers. In processing data, we are using analysis of distribution and M/G/s queuing method. The result of queuingsystem performance based on analysis and calculation shows that the queue model is (M/ G/ 8) :( FIFO/∞/∞ ). Theaverage time the customer waits in the queue is 17.17 minutes and the average service time per customer is 14.39 minutes.Customer is served for 14.39 minutes but before being served it must wait for 17.17 minutes. This condition indicatesthat waiting time is more than serving time. The system utility level at Gra PARI Banda Aceh is 95.5%, it indicatesthat utility is so high that it should be repaired to optimize the utility. The waiting time in the queue is 17.17 minutes,the waiting time in the system is 31.66 minutes, and the queue length in the system is 17 people. This will make thebottleneck of customers who will queue before being served. Based on the analysis of the new design, it is known thatthe addition of 1 server from 8 to 9 servers is the most optimal solution. The addition of 1 server can reduce the systemutility up to 84.9% and waiting time in the queue is 2.91 minutes, waiting time in the system 17.4 minutes, and the queuelength in the system 9 people. Cost analysis also shows the solution can reduce the cost of queue, so the optimal modelhas been found in this study. Therefore, optimal queuing conditions can help companies to improve service quality.

Enhanced web cache replacement policy based on data mining and RFSD scoring

By rapid growth of the Internet users and devices, the number of servers also increase simultaneously which causes the exponential increment of the internet traffic and static data. Handling these huge amounts of user requests and efficiently responding to them require high bandwidth links, powerful servers and robust equipment, which despite the availability of these requirements getting the full user satisfaction is extremely difficult and a tough challenge. In order to overcome the mentioned problem, the cache servers are being used as a suitable solution. The performance of web cache server directly depends on its replacement policies. Several cache replacement policies have been proposed in literature each having varied hit rate (HR) and byte hit rate (BHR) performances on different networks. The replacement policy proposed in this paper is a dynamic cache replacement policy which trains itself utilizing previous network logs and by exploiting the data mining clustering algorithm. Once the training step is completed, the proposed policy utilizes the normalization formulas to score each metric of the enquiries including recency, frequency, size and delay. Simulation results showed that the proposed policy has the optimum performance on different networks and it not only improved the performance of web cache server in term of HR and BHR, but also decreases the data retrieval time (Delay Ratio (DR)) of the cache servers.

Benchmarking methodology for DNS64 servers

DNS64 is an important IPv6 transition technology used in convergence with NAT64 to enable IPv6-only clients to communicate with IPv4-only servers. Several DNS64 implementations have been proposed as a solution. Their performance is an important decision factor for network operators with regard to choosing the most appropriate one among them. To that end, this article proposes a methodology for measuring their performance. The number of resolved queries per second is proposed as performance metric and a step by step procedure is given for its measurement. The design considerations behind the method are also disclosed and the performance requirements for the tester device are specified. The feasibility of our method is proven and its execution is demonstrated in two case studies, which include an empirical analysis of the tester as well as that of three open-source DNS64 implementations. The influence of the rate of existing AAAA records on the performance of the DNS64 server, as well as the influence of the cache hit rate of the DNS64 server on the performance of the DNS64 server are also measured and modeled. Our results and their precision may serve as a reference for further tests.

Optimizing web server RAM performance using birth–death process queuing system: scalable memory issue

Planning a powerful server imposes an enormous cost for providing ideal performance. Given that a server responding for web requests is more likely to consume RAM memory than other resources, it is desirable to provide an appropriate RAM capacity for optimal performance of server in congested situations. This can be done through RAM usage modeling and its performance evaluation. In the literature, modeling of RAM usage is not provided with mathematical modeling. We propose an approach to model RAM usage of such a server, based on birth–death process in this article. The model can be used to figure out an operation research problem of finding minimum RAM capacity covering intended constraints elicited from birth–death queuing system. We show how optimal RAM capacity can be obtained using our approach with an illustrative example.

Heterogeneity-Aware Collective I/O for Parallel I/O Systems with Hybrid HDD/SSD Servers

Collective I/O is a widely used middleware technique that exploits I/O access correlation among multiple processes to improve I/O system performance. However, most existing implementations of collective I/O strategies are designed and optimized for homogeneous I/O systems. In practice, the homogeneity assumptions do not hold in heterogeneous parallel I/O systems, which consist of multiple HDD and SSD-based servers and become increasingly promising. In this paper, we propose a heterogeneity-aware collective-I/O (HACIO) strategy to enhance the performance of conventional collective I/O operations. HACIO reorganizes the order of I/O requests for each aggregator with awareness of the storage performance of heterogeneous servers, so that the hardware of the systems can be better utilized. We have implemented HACIO in ROMIO, a widely used MPI-IO library. Experimental results show that HACIO can significantly increase the I/O throughputs of heterogeneous I/O systems.

HARL: Optimizing Parallel File Systems with Heterogeneity-Aware Region-Level Data Layout

Parallel file system (PFS) is commonly used in high-end computing systems. With the emergence of solid state drives (SSDs), hybrid PFS, which consists of both HDD and SSD servers, provides a practical I/O system solution for data-intensive applications. However, most existing data layout schemes are inefficient for hybrid PFS due to their unawareness of server heterogeneities and workload changes in different parts of a file. In this study, we propose a heterogeneity-aware region-level data layout scheme, HARL, to improve the data distribution of a hybrid PFS. HARL first divides a file into fine-grained, varying sized regions according to the workload features of an application, then determines appropriate file stripe sizes on servers for each region based on the performance of heterogeneous servers. Furthermore, to further improve the performance of a hybrid PFS, we propose a dynamic region-level layout scheme, HARL-D, which creates multiple replicas for each region and redirects file requests to the proper replicas with the lowest access costs at the runtime. Experimental results of representative benchmarks and a real application show that HARL can greatly improve I/O system performance, and demonstrate the advantages of HARL-D over HARL.

O efeito do treinamento na percepção do desempenho individual: uma análise baseada nos treinamentos ofertados pelo Instituto Federal do Espírito Santo (IFES)

Este artigo investiga se os treinamentos ofertados aos servidores públicos melhoram o seu desempenho no cumprimento de suas atribuições. Para tanto, optou-se por uma Instituição Pública de Ensino, cuja ampliação de suas unidades entre 2008 e 2010 resultou no aumento do quadro de pessoal e na criação da capacitação obrigatória para os novos servidores – Ambientação Institucional. A partir de um quase experimento, estimou-se o efeito do treinamento no desempenho individual dos servidores que participaram do treinamento, controlando-se por alterações no desempenho de servidores não treinados, e por outras características heterogêneas. Os resultados obtidos sugerem que o treinamento não tem impacto sobre o desempenho individual de servidores público, corroborando com a hipótese de que na ausência de mecanismo de incentivos, o treinamento não impacto desempenho.