Confimizer: A Novel Algorithm to Optimize Cloud Resource by Confidentiality-Cost Trade-Off Using BiLSTM Network

Abstract

The world is expiring a 23% annual data growth rate and is projected to have a total surplus volume of 175 Zettabytes by 2025. It imposes significant challenges for small to medium-sized businesses to allocate funds for large-size data storage. The initial large upfront and maintenance costs have made cloud storage services popular. It comes with confidentiality concerns. Encrypting data before storing it in cloud storage is the most effective solution to this challenge. Encrypting and decrypting large volumes of data allocate massive amounts of expensive resources. Storing in plain text reduces system load and expenditure but introduces confidentiality concerns. This paper proposed a Confimizer, a novel algorithm, to optimize cloud resources and reduce costs by balancing the trade-off between confidentiality and cost. It reduces the system overload by 13.75%, saving 9.20% expenditure. It saves 12.33% storage and reduces API calls by 52.99%. The Confimizer uses an optimized BiLSTM network that classifies data according to the confidentiality level by 84.00% accuracy, 76.92% precision, 74.47% recall, and 75.01 F1 score. The innovative approach, optimized BiLSTM network architecture, and outstanding performance of the Confimizer make it a unique and effective cloud resource optimization algorithm.

Keywords

• Cloud computing
• resource optimization
• data confidentiality
• optimization
• deep learning