A hybrid convolutional neural network with long short-term memory (HCNN-LSTM) model based Edge System Recommendation(ESR) for cloud service providers

Abstract

Massive entities with significant state denaturation exist in the physical world, and users urgently want real-time and intelligent entity information acquisition. Now a days hand held devices and Electronic Device usage are increased rapidly. For effective cloud service edge systems are used. As rapid growth of internet and hardware technologies many edge providers are in the market. For Cloud Service Providers (CSP) it will be a chaos to choose optimised and reliable edge system for their usage. A hybrid convolutional neural network with Long Short-Term Memory (HCNN-LSTM) based Edge System Recommendation (ESR) Algorithm for Cloud Service Providers. An entity identification technique suitable for the edge is suggested to enhance the precision and responsiveness of entity state data search. It conducts accurate entity identification using the HCNN-LSTM while considering entity feature information. Even for new CSP optimised Edge System will be easily allocated. For this Recommendation Algorithm a total of 3 Servers and 25 Edge Systems with varied configuration are tested. 1000 GB hybrid data is used for transaction. The dataset collected from Kaggle online web-based repository. This proposed algorithm proves 95% more effective than existing techniques.

Keywords

• Cloud servers
• Low configuration
• Multiple edge systems
• Cloud service Providers
• Edge system recommendation (ESR)