Performance Evaluation of a Hyper Converged Architecture Based on Microsoft Storage Spaces

• Main Authors: LU,CHIA-YU, 呂嘉祐
• Other Authors: LIN,TSUNG-WU
• Format: Others
• Language: zh-TW
• Published: 2018
• Online Access: http://ndltd.ncl.edu.tw/handle/e7gk2r

碩士 === 東吳大學 === 資訊管理學系 === 106 === The resources required by the current IT infrastructure can be almost completely provided by the server with virtualization platform software. Not only Software Defined Compute (SDC) related virtual processors and memory applications, but also Software Defined Storage (SDS) is an environment that can be configured on bare metal servers or be integrated into a server virtualization platform, providing storage requirement in the form of a Hyper-Converged Infrastructure. It is a hot topic for enterprises reducing the cost of building a storage environment by the server host and embedded disk resources, software-defined approach, to create a high-availability and extensibility of the software-defined storage architecture. The Microsoft Windows Server 2016 provides the Storage Spaces Direct (S2D), a hyper-convergence architecture distributed storage system that is directly embedded in the core of the operating system and adds enterprise-level non-transient write caches in response to performance requirements. (Non-Volatile Write Cache, NVME) SSDs are used as data cache I/O exchanges. We will analyze and explain the storage operation efficiency and performance of different functional applications and provide enterprises with an assessment of the actual production environment and planning reference.