Sparse Regression Model to Predict a Server Load for Dynamic Adjustments of Server Resources

Sparse Regression Model-Based Relearning Architecture for Shortening Learning Time in Traffic Prediction

“…LASSO also has characteristics that it achieves agile prediction processing and improvement of memory consumption by making a short list of explanatory variables. Therefore, our approach is to use LASSO [5] to reduce learning data and shorten learning time of machine learning.…”

“…In [5], we formulate a LASSO based regression analysis model. Our model regards the values of server loads monitored every one minute in the time period from t 1 to t 2 as explanatory variables, and predicts the average value of server loads in the period from t 3 to t 4 .…”

“…To reduce the errors as much as possible, trying all possibilities of time periods requires too much learning time. Thus, previously, we proposed an effective algorithm to determine the time period targeted for explanatory variables, of which the detail is described in [5]. Our algorithm determines the time period targeted for explanatory variables by iterative learning processes.…”

“…Therefore, reducing data is one of the most common challenges in machine learning. We firstly present our least absolute shrinkage and se-Copyright c 2021 The Institute of Electronics, Information and Communication Engineers lection operator (LASSO) based sparse regression model to predict a virtual server load [5]. In our previous work [5], we dealt with the sparsity of real-life dataset of number of access to an operational Web server which is targeted as a virtual server load, and evaluated the effectiveness of our model from the perspective of prediction error.…”

Intelligent and Agile Control of Edge Resources for Latency-Sensitive IoT Services

“…Among them, the decision tree, k-nearest neighbor, gradient boosting, and linear regression techniques have been employed to predict the penalty for service-level agreement violation of CPU-intensive database applications and allocate the appropriate amount of resources [11]. The lasso regression model with regularization was used to predict server workloads in [13]. In [12], autoregressive models for CPU resource demand prediction and allocation were suggested to virtualized servers operating in enterprise datacenters.…”