Providing dynamic resources is based on the virtualization features of the cloud environment. Cloud computing as an emerging technology uses a high availability of services at any time, in any place and independent of the hardware. However, fault tolerance is one of the main problems and challenges in cloud computing. This subject has an important effect on cloud computing, but, as far as we know, there is not a comprehensive and systematic study in this field. Accordingly, in this paper, the existing methods and mechanisms are discussed in different groups, such as proactive and reactive, types of fault detection, etc. Various fault tolerance techniques are provided and discussed. The advantages and disadvantages of these techniques are shown on the basis of the technology that they have used. Generally, the contributions of this research provide a summary of the available challenges associated with fault tolerance, a description of several important fault tolerance methods in the cloud computing and the key regions for the betterment of fault tolerance techniques in the future works. The advantages and disadvantages of the selected articles in each category are also highlighted and their significant challenges are discussed to provide the research lines for further studies.

中文翻译：

---

云计算容错架构的全面系统研究

提供动态资源是基于云环境的虚拟化特性。云计算作为一种新兴技术，在任何时间、任何地点使用高可用性的服务，并且独立于硬件。然而，容错是云计算的主要问题和挑战之一。该课题对云计算有重要影响，但据我们所知，该领域尚无全面系统的研究。因此，在本文中，现有的方法和机制在不同的组中进行了讨论，例如主动和被动，故障检测类型等。提供和讨论了各种容错技术。这些技术的优点和缺点是根据它们使用的技术来显示的。一般来说，本研究的贡献总结了与容错相关的可用挑战，描述了云计算中几种重要的容错方法以及在未来工作中改进容错技术的关键领域。还突出了每个类别中所选文章的优缺点，并讨论了它们面临的重大挑战，为进一步研究提供研究方向。