Comparative Study of Neural Networks Algorithms for Cloud Computing CPU Scheduling

Hicham, Gibet Tani; Chaker, El Amrani; Elaachak, Lotfi

doi:10.11591/ijece.v7i6.pp3570-3577

Cited by 10 publications

(9 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The learning rate, β 1 , β 2 , and of the Adam optimizer are set according to [38]. We iterate each FIGURE 2: A simplified Neural Network block diagram [46] of our models 100 times.…”

Section: A Proposed Architecturementioning

confidence: 99%

“…The network trains the data to identify the patterns present within the data and adjust the associated weights to predict the output for a brand new set of similar data. Weight and bias in each neuron and edge are optimized using the backpropagation technique, which employs gradient descent [46]. The last part of the BHyPreC model incorporates a fully connected dense neural network having a single output layer neuron to calculate the final prediction.…”

Section: B Neural Networkmentioning

confidence: 99%

See 1 more Smart Citation

BHyPreC: A Novel Bi-LSTM Based Hybrid Recurrent Neural Network Model to Predict the CPU Workload of Cloud Virtual Machine

Karim

Maswood²,

Das

et al. 2021

IEEE Access

View full text Add to dashboard Cite

With the advancement of cloud computing technologies, there is an ever-increasing demand for the maximum utilization of cloud resources. It increases the computing power consumption of the cloud's systems. Consolidation of cloud's Virtual Machines (VMs) provides a pragmatic approach to reduce the energy consumption of cloud Data Centers (DC). Effective VM consolidation and VM migration without breaching Service Level Agreement (SLA) can be attained by taking proactive decisions based on cloud's future workload prediction. Effective task scheduling, another major issue of cloud computing also relies on accurate forecasting of resource usage. Cloud workload traces exhibit both periodic and non-periodic patterns with the sudden peak of load. As a result, it is very challenging for the prediction models to precisely forecast future workload. This prompted us to propose a hybrid Recurrent Neural Network (RNN) based prediction model named BHyPreC. BHyPreC architecture includes Bidirectional Long Short-Term Memory (Bi-LSTM) on top of the stacked Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU). Here, BHyPreC is used to predict future CPU usage workload of cloud's VM. Our proposed model enhances the non-linear data analysis capability of Bi-LSTM, LSTM, and GRU models separately and demonstrates better accuracy compared to other statistical models. The effect of variation of historical window size and training-testing data size on these models is observed. The experimental result shows that our model gives higher accuracy and performs better in comparison to Autoregressive Integrated Moving Average (ARIMA), LSTM, GRU, and Bi-LSTM model for both short-term ahead and long-term ahead prediction.INDEX TERMS Cloud Computing, deep learning, recurrent neural network, time series analysis, virtual machine, workload prediction.

show abstract

“…The learning rate, β 1 , β 2 , and of the Adam optimizer are set according to [38]. We iterate each FIGURE 2: A simplified Neural Network block diagram [46] of our models 100 times.…”

Section: A Proposed Architecturementioning

confidence: 99%

Section: B Neural Networkmentioning

confidence: 99%

BHyPreC: A Novel Bi-LSTM Based Hybrid Recurrent Neural Network Model to Predict the CPU Workload of Cloud Virtual Machine

Karim

Maswood²,

Das

et al. 2021

IEEE Access

View full text Add to dashboard Cite

show abstract

“…A neural network consists of an input layer, zero or more hidden layer, and an output layer. Information on neurons will be propagated through each layer, starting from the input layer, all the way to the output layer [14] [15]. The propagation is done by calculating the weighted sum on each neuron, which then used as the input value for the activation function used.…”

Section: A Neural Network and Backpropagationmentioning

confidence: 99%

“…The backpropagation helped a neural network to learn the relationship between variables, without explicitly defining the mathematical function that defines that relationship [17]. A neural network trained using backpropagation is based on gradient descent [14]. In the backpropagation algorithm, each neuron in the hidden and output layer will process its own inputs using the sum product of each neuron's input value and weight of each neuron, respectively, which then processed through an activation function (most popular used is sigmoid activation function) [17].…”

Section: A Neural Network and Backpropagationmentioning

confidence: 99%

Earthquake Magnitude and Grid-Based Location Prediction using Backpropagation Neural Network

Priambodo¹,

Mahmudy²,

Rahman³

2020

KEDS

View full text Add to dashboard Cite

Even though it is inevitable, but it can be anticipated to minimize damage and casualties. Past research has been conducted to predict the level of impact caused by earthquakes in real-time [2]. One of those past research is about an early earthquake warning system (EEWS), which will give an alert when it detects an earthquake [3]. Numerous architectures and algorithms have been developed in those studies. Various research is the utilization of neural network trained with backpropagation algorithm and optimized using Levenberg (LOM) to predict hypocenter location, moment magnitude, and the expansion of the earthquake [4], modification of LOM to minimize error on EEWS [3], and utilization of neural tree to predict P and S waves [5]. Until now, to the knowledge of the author, the study of using a backpropagation (BP) algorithm to predict earthquake magnitude and grid-based location in Indonesia has not been conducted yet. This algorithm is chosen because it has been proven to perform well in broad types of problems, such as regression, pattern recognition, and prediction [6][7][8][9]. In this paper, the study aims to measure the performance of neural network trained using backpropagation algorithm in predicting earthquakes magnitude and grid-based location based on earthquakes magnitude and location data in Indonesia recorded from 2000 to 2019.

show abstract

“…The needs of learn ing outcomes for co mputer engineering undergraduate students a re about four levels of the CPU area: processor abstraction, CPU organization, microcode, and its implementation [27]. Some free softwares are reco mmended as another alternative to improve the quality of the teaching.…”

Section: Literature Reviewmentioning

confidence: 99%

CPU implementation using only logisim simulator to achieve computer architecture learning outcome

Ichsan

Kurniawan

2020

Bulletin EEI

View full text Add to dashboard Cite

This paper proposes the design, implementation, and evaluation of an 8-bit CPU architecture in computer organization and architecture (COA) course for Undergraduate Computer Engineering using only Logisim simulator. The main advantage of using one simulator is the elimination of inefficiency so that students can concentrate more on the course’s content without spending their time just to learn how to use each different type of simulators. To achieve that, a simple CPU architecture is pre-designed and be implemented in Logisim simulator. According to previous researches, Mic-1 CPU architecture is chosen because it is the simplest one and can be easily built using many simple logic gates already existed in the Logisim. To evaluate the desired outcome, students are separated into two different groups. Each group used different type of learning media and material and then their examination scores and satisfaction are compared to each other. Every student belongs to the group that only used Logisim simulator obtained higher score and more than 50% are satisfied with the new learning process and material. Thus, it envisaged that this method will made the delivering learning process in COA course be more efficient than what has happened so far.

show abstract

Comparative Study of Neural Networks Algorithms for Cloud Computing CPU Scheduling

Cited by 10 publications

References 30 publications

BHyPreC: A Novel Bi-LSTM Based Hybrid Recurrent Neural Network Model to Predict the CPU Workload of Cloud Virtual Machine

BHyPreC: A Novel Bi-LSTM Based Hybrid Recurrent Neural Network Model to Predict the CPU Workload of Cloud Virtual Machine

Earthquake Magnitude and Grid-Based Location Prediction using Backpropagation Neural Network

CPU implementation using only logisim simulator to achieve computer architecture learning outcome

Contact Info

Product

Resources

About