A survey of Cloud Security

A Survey of Cloud Security

Idea

云计算的安全在大数据发展的历程中逐渐受到重视,数据管理难度增大的同时,也显现出了管理下原本传统模式可能预见不到的可能攻击手段,因此数据IO扩容的情况下数据安全也需要加深研究。

那么,现在的云计算场景中,我们需要探讨关于安全的问题的话,首先需要将这个问题拆解成小问题,确定和明确目前最前沿已经成熟的技术和可能出现的攻击手段,以及尚未成熟,值得研究的领域。

  • 云计算场景下,相比于传统场景,暴露出的新的安全问题有哪些?
  • 云计算相比传统场景,基于什么样的设备和系统的变化才导致了新的安全问题?
  • 云计算场景下,新的安全问题已经提出了什么样的解决方案?是否存在还不成熟的安全威胁领域?是否存在还未被解决的安全问题?
  • 云计算场景必须满足高负载低延迟的需要,那么现有的安全方案是否会过分削弱原本的方案的数据传输带宽?是否在实现上不具备使用价值?

以上大致是我暂时想到的一些理论上的问题,我需要根据上述问题进行简单的但必须是全面的(也许并不需要,假设我发现了一个空缺问题,只要我确定我能解决,且没有其他更大的突变问题,那我就试试看)探究当前学术界对于云计算场景的安全问题研究发展现状,以满足我短暂三年内快速发论文然后毕业的需求。

Introduction

本章主要目的是为了引入云计算技术背景,以便后续技术研究报告的学术名词不会将我搞混,但我未必需要马上把这章写好才能写后续的内容,我可以一步步的从最新领域下确认各个学术名词,并在此做个标记。

​ 云计算所面临的数据安全问题,其中最尖锐的应当属细粒度属性的访问控制,进一步深入考察,其中密钥泄漏读写权限灵活性不足以及属性撤销开销过大,是威胁到数据安全服务体系质量的几个关键方面。

Technology Frontier

以下论文技术相关顺序仅代表我个人阅读顺序

​ 为了实现有效云计算框架之下的安全服务,目前比较常见的数据访问控制策略 包括基于身份认证的访问控制以及基于属性加密的数据访问控制两种。^[1]^

​ 其中前者主要是在数据加密的基础之上引入可以信任的第三方,来对用户身份进行考察验证,从而确保公钥安全。

​ 但是后者在当前的云环境中则更为常见,这是一种将数据传输过程中角色与权限绑定修改为属性与权限相结合的控制机制。与比较传统的数据访问控制相比,此种控制方案可以改善细粒度不足的问题,尤其符合当前云背景之下一对多的访问要求。

​ 当前属性加密访问控制领域中,密文策略属性基加密方案(CP-ABE,Ciphertext Policy Attribute Based Encryption)比较常见。^[1]^

在此种方案之下,用户属性集合负责定义和描述用户解密密钥,而数据属主定义的访问结构则用于确定数据密文的生成方式。在此框架下共有四个角色,即认证中心、云服务商,数据拥有者以及用户。其中认证中心负责为数据拥有者和用户分别提供公钥和私钥,数据拥有者将数据送达云服务商,并且约定密文访问策略,而用户则面向云服务商提出访问请求,并且当符合对应的属性条件就可以执行解密。

这个过程对于用户来说比较简单,为整个云环境带来的运算量也十分有限,尤其是当云规模不断扩大的时候,此种策略仍然有着良好生命力,但是仍然存在者属性撤销以及密钥泄漏等风险需要解决

CP-ABE

​ CP-ABE计算框架之下,对于加密方面而言,是首先由授权中心生成系统公钥以及主密钥,而后将这些密钥进行分发,公钥发送给数据提供方,允许其使用公钥和访问结构来对数据进行加密并且上传到云端。而在解密方面,授权中心能够依据用户属性生成私钥配给给用户,用户可以用私钥来对从云平台中获取到的加密数据进行解密。私钥本身是和用户属性绑定的,如果用户属性不满足则无法实现解密。

进一步CP-ABE工作过程中的细节进行考察,发现有几个方面的问题必须引起关注。例如数据用户可以在CP-ABE的支持下实现对于云端数据的读取,但是如果需要修改或者写入操作,则必须设置对应的控制访问权限参数才能落实。而访问控制中对于密钥的保护,还必须从明文和密文两个角度同时出发才能落实,并且还需要在同时对加密算法本身要求的运算能力做出考虑,过于复杂的运算会导致系统整体工作效率的下降。除此以外,当用户属性发生变更的时候,诸如撤销属性或者权限改变等,还必须能够做到及时撤回私钥,落实数据的保护。而这个过程中的效率,就成为关系到云环境中数据安全的关键所在。

Reference

[1] 李佳恩.云计算背景下数据安全加密算法探究[J].中国信息化,2021,(11):62-63.

摘要:云计算领域中,安全一直都是众人关注的重点,而与安全一同出现的效率问题同样不容忽视。本文针对相关问题展开分析,提出了双层加密的算法设想,并且就其应用特征展开了分析。

A survey of Storage Virtualization

A Survey of Storage Virtualization

Introduction

With the development of information technology and data processing technology, especially the rise of 5G communication technology, the traditional information system can no longer meet the demand for efficient and high-capacity storage with low latency. Traditional information processing and storage systems are cumbersome, have limited storage size and have low application rates. In order to optimize the computer storage function, storage virtualization technology has emerged. After more than a decade of development, storage virtualization has been gradually accepted by people, improving the storage performance and security performance of computer systems.

The challenges of adopting traditional storage

When the traditional RAID group storage mode is deployed, the configuration planning and later maintenance of the business is extremely complicated. In the 5G environment, the storage volume is bound to surge, and when the number of hard disks reaches a certain level, not only is the configuration process more prone to errors, but the later adjustments will be more complex, which will bring a lot of management and maintenance costs.

There is also a very serious problem, the use of traditional storage will lead to performance risks, system operation and recovery speed are difficult to meet the requirements of 5G low latency. In recent years, the growth of hard disk capacity, and hard disk read and write rate growth is extremely mismatched, hard disk capacity has grown nearly 100 times, while read and write performance growth of even less than 2 times. The direct result of this asymmetric growth is the use of physical hard disk as a unit to do the traditional way of RAID inefficiency and increasingly poor reliability.

The traditional storage model is also unable to meet the requirements of 5G services in backup recovery operations.

Storage virtualization and its benefits

Storage virtualization is the process of converting traditional computer hardware data storage into virtual storage. Storage virtualization integrates a single function within a system to enhance the comprehensiveness of the system. Storage virtualization technology is a means of virtualizing storage, which virtualizes complex and diverse real-world technology devices in an abstract way and enables good control of real-world physical devices. From the user’s point of view, storage virtualization is the original disk, hard disk storage data technology for virtualization, the application of all the data stored through virtualization, the user no longer have to consider the storage location and storage security. From the enterprise manager’s point of view, storage virtualization stores all enterprise data and information in the virtualized storage pool, and integrates information management, so that enterprise managers can use information more quickly and efficiently. This is also the case with virtual tape libraries, which are often used by enterprises today.

With the use of storage virtualization technology, a cloud storage space with huge capacity can be developed separately from the hardware storage space of the host computer, and users can upload and store many types of information, which solves the problem of incompatibility between different information and storage devices and can save a lot of hardware acquisition costs for users.

At the same time, it can also greatly improve the security of information data, generally speaking, the information stored in hardware facilities, if there is an accidental power failure and other conditions, it is easy to lead to abnormalities in the storage area and lead to information loss, which is bound to be a huge loss for users, while the cloud space is not affected by these factors, even if the information is lost due to the user’s wrong operation, it can also be recovered through the file recovery technology mechanism in the cloud platform, reducing the loss caused by operational errors.

Storage Virtualization Technology

Storage virtualization technologies can be divided into host-based storage virtualization, device-based storage virtualization, and network-based storage virtualization according to the principle of implementation.

Host-based Storage Virtualization

The control software of this data storage method is installed on the host computer, so that the data stored in the subsystem can be retrieved and viewed directly through the host computer. A host can be connected to several agent systems, and each agent system can share a certain amount of data storage capacity, reducing the pressure on the host data storage. At the same time, the host virtualization technology is easy to implement, and the equipment cost is low, which has a high cost performance.

However, this virtual storage method also has shortcomings, such as scalability is not strong, the efficiency of the use of internal storage data will be affected by the operational stability of the host.

In addition, there may be data protection barriers between the various agent systems, and even access rights may be restricted, which will also affect the application value of virtual storage.

Device-based Storage Virtualization

The storage devices mentioned here refer to the storage modules distributed in the virtual terminal. Virtual storage based on storage devices also has its own advantages, such as easier management, with the help of information integration technology, you can dynamically grasp the storage capacity of each storage module, remaining capacity and other basic information on the computer.

However, as an independent component of the computer system, the storage device can provide relatively limited data storage capacity, and with the increase in the number of storage modules inside the computer, there will be signal crosstalk between each other and other problems, which will also have a negative impact on the security of stored data.

Network-based Storage Virtualization

The Internet can provide a dedicated server for virtual computer storage, and the server implements standard operating systems, such as the commonly used Windows, Linux, etc., ensuring the security of the virtual storage environment and the confidentiality of data information.

The advantage of this storage method is that there is no need for individuals or units to purchase storage equipment, but the Internet provider provides a unified shared server, and any enterprise or individual can use the virtual server with a certain storage capacity after the main audit, so it has certain advantages in terms of storage capacity, data security and cost.

Summary

Computer storage virtual technology occupies the computer storage market with its great advantages. The application of this technology not only facilitates people’s work in the computer field, but also provides space for the development of large-scale storage of information data in the era of big data. In the future development, we need to rely on virtualization software to support the development of storage virtualization technology, improve the performance of server storage, so as to better meet the needs of data storage in the era of big data.

References

[1]Guanglei Zhu. Cloud Storage Trend of Technology Development and Application[J]. Computer Knowledge and Technology, 2011, 7(11):2561-2562.

[1]Xiujuan Tian. Discussion on Storage Virtualization Technology[J]. Information and Computer(Theoretical Version),2020,32(02):15-17.

[3]Qingquan Dong, Qianjun Wu, Yuhang Cheng. Research on Cloud Computing Based on Storage Virtualization in Data Center[C]. Science and Engineering Research Center. Proceedings of 2019 International Conference on Mathematics, Big Data Analysis and Simulation and Modeling(MBDASM 2019).Science and Engineering Research Center:Science and Engineering Research Center,2019:217-219.