Enhanced Virtual Machine Placement in Cloud Data Centers: Combinations of Fuzzy Logic with Reinforcement Learning and Biogeography-Based Optimization (BBO) Algorithms

The process of mapping Virtual Machines (VMs) to Physical Ma- chines (PMs), which is defined as VM placement, affects Cloud Data Centers (DCs) performance. To enhance the performance, optimal placement of VMs regarding conflicting objectives has been proposed in some research, such as Multi-Objective VM reBalance (MOVMrB) and Reinforcement Learning VM reBalance (RLVMrB) in recent years. The MOVMrB algorithm is based on the BBO meta-heuristic algorithm and the RLVMrB algorithm inspired by reinforcement learning, which in both of them the non-dominance method is used to evaluate generated solutions. Although this approach reaches accept- able results, it fails to consider other solutions which are optimal regarding all objectives, when it meets the best solution based on one of these objectives. In this paper, we propose two enhanced multi-objective algorithms, Fuzzy- RLVMrB and Fuzzy-MOVMrB, that are able to consider all objectives when evaluating candidate solutions in solution space. All four algorithms aim to balance the load between VMs in terms of processor, bandwidth, and memory as well as horizontal and vertical load balance. We simulated all algorithms using the CloudSim simulator and compared them in terms of horizontal and vertical load balance and execution time. The simulation results show that Fuzzy-RLVMrB and Fuzzy-MOVMrB algorithms outperform RLVMrB and MOVMrB algorithms in terms of vertical load balancing and horizontal load balancing. Also, the RLVMrB and Fuzzy-RLVMrB algorithms are better in execution time than the MOVMrB and Fuzzy-MOVMrB algorithms.

Analisa Pembagian Penggunaan Bandwidht Untuk Jaringan Internet Pelanggan Mnc Play Area Jakarta Dengan Menggunakan Teknologi Link Aggregation Dan Load Balance

<p>Dalam sebuah perusahaan penyedia layanan jaringan internet, di tuntut untuk memiliki<br />kulitas jaringan <em>high availability</em> dan <em>high reliability</em>. Dalam mendukung kualifikasi<br />jaringan internet terbaik maka salah satunya perlu menjaga kestabilan bandwidth<br />internet pelanggan. Salah satunya menjaga kepadatan traffic data pada jaringan internet.<br /><em>Congestion</em> merupakan pengumpulan paket data yang melebihi kapasitas bandwidth<br />yang tersedia padasebuah link, hal ini memberikan dampak penurunan kinerja jaringan<br />internet. Permasalahan di atas dapat di selesaikan salah satunya dengan menerapkan<br />penggunaan metode <em>link aggregation</em> dan teknologi <em>load balance</em>. Pada penelitian ini<br />penulis menerapkan metode <em>link aggregation</em> yang digunakan dalam merancang<br /><em>redundancy link</em> dan <em>load balance</em> dalam membuat skema membagi traffic untuk<br />penyeimbang beban berdasarkan <em>src address</em> dan <em>src port</em>, serta <em>dst address</em> dan <em>dst</em><br /><em>port.</em> Dari hasil pengujian penggunaan <em>link aggregation</em> dan <em>load balance</em> pada<br />penenlitian ini dapat membantu menjaga kestabilan throughput dengan nilai pengukuran<br />sebesar 10GB dengan standart MTU (<em>Maximum Transmission Unit</em>) yang menandakan<br />hasil test pengukuran paket data terbesar yang dapat di transmisikan melalui sebuah<br />jaringan.</p><p><br /><em><strong>Keyword</strong> : link aggregation, load balance, congestion, MTU</em></p>

On the Transformation Optimization for Stencil Computation

Stencil computation optimizations have been investigated quite a lot, and various approaches have been proposed. Loop transformation is a vital kind of optimization in modern production compilers and has proved successful employment within compilers. In this paper, we combine the two aspects to study the potential benefits some common transformation recipes may have for stencils. The recipes consist of loop unrolling, loop fusion, address precalculation, redundancy elimination, instruction reordering, load balance, and a forward and backward update algorithm named semi-stencil. Experimental evaluations of diverse stencil kernels, including 1D, 2D, and 3D computation patterns, on two typical ARM and Intel platforms, demonstrate the respective effects of the transformation recipes. An average speedup of 1.65× is obtained, and the best is 1.88× for the single transformation recipes we analyze. The compound recipes demonstrate a maximum speedup of 1.92×.

Effect of Relative Humidity in Air on the Transmission of Respiratory Viruses

Viral respiratory infections have plagued mankind over its known history. Unfortunately, there has been a lack of meaningful progress in preventing the spread of viral respiratory infections globally. The central dogma appears to be that viruses are the villains. This framing focuses on a viral load balance (VLB) in the air. It follows that physical dilution through various means have been the primary focus of attempts to reduce the spread of infections. The problem of obesity provides a good example of how paradigm blindness can slow down progress in a field. Obesity has been framed as an energy balance disorder that blames overeating and lack of exercise for weight gain. Reframing obesity as a disorder of fat metabolism and storage caused by the quantity and quality of carbohydrates in the diet, referred to as the carbohydrate-insulin model (CIM), opened an alternative line of questioning with a testable hypothesis. Similarly, we postulate an alternative way to frame the spread of viral respiratory infections that would lead to new insights and potentially new ways to prevent infections. It has long been recognized that viral respiratory infections show a pronounced seasonal variation, referred to as seasonal forging, such that they increase in the winter but decrease or virtually disappear in the summer. In temperate regions, people spend over 90% of their time indoors. This is, therefore, where most respiratory infections are expected to occur. Evidence has been accumulating for decades on the strong correlation between variations in indoor relative humidity (RH) and variations in infection rates. Within a RH Goldilocks zone of 40%-60%, encapsulated viruses like influenza and SARS are optimally inactivated outside the infected host. Below 40% and above 80%, viruses can survive for extended periods in the air or on surfaces. This may explain in part the seasonality of infections as the indoor level of RH in winter is typically about 20% and above 40% in summer in temperate regions. However, the mechanism for the inactivation at midrange RH (in summer) is not well understood. This paper offers a hypothesis that could explain these observations. We have demonstrated that H2O2 and other reactive oxygen species (ROS) are formed spontaneously at the water-air interface of pure water microdroplets. Using only water and a nebulizing gas in the presence of oxygen, we have demonstrated the significant disinfectant potential of pure water microdroplets caused by the activity of H2O2 and other ROS. We postulate that spontaneous H2O2 and ROS formation in viruses containing exhaled microdroplets have a similar virucidal effect at mid-range RH. The droplet evaporation rate is sufficient to concentrate the solutes and provide enough time for reactions to occur at significantly higher rates than in bulk solutions. The concentration of H2O2 has also been shown to be positively correlated to RH. In addition, several other ROS/RNS may be present or formed through interactions with H2O2 that may act as even more effective virucide disinfectants to inactivate the virus. Below RH 40% evaporation happens too rapidly for these reactions to make an impact before the droplet is desiccated, and above RH 80% the solutes remain too diluted. Rapid inactivation of viruses at midrange RH may therefore play a greater role in preventing infections than physical dilution of virus load in the air through excessive mechanical ventilation. Similar to obesity, we suggest that a new paradigm that considers virus infectivity outside the host rather than the virus load balance in the air alone could greatly contribute to our understanding of respiratory infections. The proposed new “Relative Humidity Infectivity” RHI paradigm could explain the causal mechanisms underlying seasonal respiratory infections. This can point to better prevention strategies that avoid further distortion of our indoor environment and create conditions within which humans can thrive and be optimally protected. We need more focus on testing the various hypotheses and more data to determine which of the two paradigms will lead us in the right direction or how to use the best of both in an optimal combination. The stakes cannot be higher, and the potential for eradicating future viral respiratory pandemics with nature-based solutions may be right under our noses, literally.

Intelligent Classification Method of Low Occupancy Big Data Based on Grid Index

In order to enhance the load balance in the big data storage process and improve the storage efficiency, an intelligent classification method of low occupancy big data based on grid index is studied. A low occupancy big data classification platform was built, the infrastructure layer was designed using grid technology, grid basic services were provided through grid system management nodes and grid public service nodes, and grid application services were provided using local resource servers and enterprise grid application services. Based on each server node in the infrastructure layer, the basic management layer provides load forecasting, image backup, and other functional services. The application interface layer includes the interfaces required for the connection between the platform and each server node, and the advanced access layer provides the human-computer interaction interface for the operation of the platform. Finally, based on the obtained main structure, the depth confidence network is constructed by stacking several RBM layers, the new samples are expanded by adding adjacent values to obtain the mean value, and the depth confidence network is used to classify them. The experimental results show that the load of different virtual machines in the low occupancy big data storage process is less than 40%, and the load of each virtual machine is basically the same, indicating that this method can enhance the load balance in the data storage process and improve the storage efficiency.

Resource Management and 3D Deployment in Power-Efficient Multi-UAV Networks

In this paper, the scenario in which multiple unmanned aerial vehicles (UAVs) provide service to ground users is considered. Under the condition of satisfying the minimum rate per user and system load balance, the user association, bandwidth allocation and three dimensional (3D) deployment of multi-UAV networks are optimized jointly to minimize the total downlink transmit power of UAVs. Since the problem is hard to solve directly, it is decomposed into three sub-problems, and then the problem is solved by alternating iteration algorithm. First, when the UAV’s location is determined, a modified K-means algorithm is used to obtain balanced user clustering. Then, when the user association and UAV’s 3D deployment are determined, the convex optimization method is used to obtain the optimal bandwidth allocation. Finally, when the user association and optimal bandwidth allocation are determined, a modified differential evolution algorithm is proposed to optimize the 3D location of the UAVs. Simulation results show that the proposed algorithm can effectively reduce the transmit power of UAVs compared with the existing algorithms under the conditions of satisfying the minimum rate of ground users and system load balance.

IMPLEMENTASI LAOD BALANCING PADA WEB SERVER MENGGUNAKAN APACHE DENGAN SERVER MIRROR DATA SECARA REAL TIME

ABSTRAKPT.Trimitra Data Teknologi adalah perusahaan yang yang bergerak dalam bidang teknologi dan informasi, website menjadi salah satu cara jembatan komunikasi antara client dan perusahaan. Banyaknya client yang mengakses membuat beban sebuah web server dalam perusahaan menjadi berat dan menimbulkan masalah yaitu down nya server yang membuat client sulit untuk mengakses website perusahaan. Untuk membantu mengatasi masalah yang terjadi diterapkannya metode load balancing dengan algoritma request counting algorithm dimana bertujuan untuk membagi beban secara merata dalam web server dan memperkecil waktu respon antara client dan server, beban terbagi dengan anggota server yang terdaftar dalam server load balancing. Dengan penerapan metode load balancing maka kerja server akan menjadi lebih maksimal karena adanya sistem high availability dimana saat salah satu server mati maka kerja server akan diambil alih oleh server yang lain. Selain metode load balance penerapan sistem dengan server mirror yang dilakukan dapat membantu memaksimalkan metode load balance karena adanya replikasi otomatis antara web server yang menjadi anggota load balance baik konten website ataupun database. Hasil yang terjadi adalah web server perusahaan akan menjadi sistem yang mampu bekerja secara baik saat melayani client dalam hal layanan web server karena beban terbagi dengan baik dan kecilnya waktu respon sehingga tidak adanya kesulitan client untuk mengakses website perusahaan.Kata kunci : Load Balance, Web Server, Mirror Server.

Simulation analysis of cooperative motion fuzzy control of distributed lifting units

In the technology of hydraulic lifting system, it is not only necessary to ensure that the displacement and velocity accuracy of each hoist reach a certain value, but also to ensure that under the control of load balance to make each hoist smooth lift. In the conventional method, the PID control method can realize the synchronization of the function. However, the system cannot be controlled and adjusted in real time during the control parameter period, resulting in instability and uncertainty of the system. Aiming at this problem, this paper adds the fuzzy adaptive controller to carry out the master-slave control of the system. AMESim and MATLAB co-simulation were used to model the overall model of the hydraulic system. At the same time, the pressure compensator and variable throttle port model in the hydraulic reservoir were selected to build. The pressure compensator is used to keep the pressure difference of the throttle orifice constant, so as to complete the control and design of the hydraulic lifting system. Finally, the simulation results obtained not only can effectively improve the instability of the hydraulic lifting process, but also greatly improve the operation speed of the system.