基于DRBD双控制器磁盘阵列系统控制与管理
发布时间：2018-12-16 08:05
【摘要】：磁盘阵列控制器作为磁盘阵列的核心部件，它的性能高低已经成为影响磁盘阵列系统整体性能的关键因素，与之相应的磁盘阵列存储系统的故障单点开始逐渐向控制器端转移，通过控制器的冗余机制，保证了在控制器端出现单点故障的情况下，系统不会瘫痪，，仍能对外提供不间断服务。然而，这种方式只是避免了应用的单点故障，如今对于数据的安全性有着更高要求，尤其是某些部门需要提供7*24的不间断服务，磁盘阵列系统不仅需要有控制端的冗余，还应该充分保证数据存储端的可靠性。 基于DRBD（Distributed Replicated Block Device）的双控制器磁盘阵列是一套双机双柜的解决方案，提供基于iSCSI(Internet Small Computer System Interface)的网络存储服务。通过逻辑卷管理实现了设备级别的存储虚拟化，使对磁盘设备的管理更加灵活；除了采用两台控制器同时提供服务保证设备的利用率与避免控制器的单点故障外，还利用DRBD的数据镜像功能实现了存储数据的冗余，实现了控制器端与存储设备端的双重容错的功能，提高了系统的可用性。系统通过将存储设备划分为普通卷与高可用卷来提供相应级别的存储服务，满足用户多样性的需求。 系统的控制管理采用了B/S(Browser/Server)的架构，包括前台的Web管理界面与后台的守护进程，实现了对阵列远程可视化管理，有效的解决了双机系统故障的监控，处理与恢复，测试结果表明系统达到了设计的要求。
[Abstract]:As the core component of the array, the performance of the array controller has become the key factor to affect the overall performance of the array system. The corresponding single point of the disk array storage system starts to transfer to the controller side gradually. The redundancy mechanism of the controller ensures that the system will not be paralyzed and can still provide continuous service in the case of single point failure at the controller end. However, this approach only avoids a single point of failure in the application, and today there are higher requirements for data security, especially in some departments that need to provide uninterrupted service of 724, and disk array systems not only need redundancy on the control side, The reliability of the data store should also be fully guaranteed. Dual controller disk array based on DRBD (Distributed Replicated Block Device) is a dual computer and double cabinet solution, which provides network storage service based on iSCSI (Internet Small Computer System Interface). The storage virtualization of device level is realized through logical volume management, which makes the management of disk devices more flexible. In addition to using two controllers to guarantee the utilization of the equipment and avoid the single point of failure of the controller, the data mirror function of DRBD is used to realize the redundancy of the stored data. The dual fault-tolerant function of controller and storage device is realized, and the availability of the system is improved. By dividing storage devices into ordinary volumes and high availability volumes, the system provides storage services at the corresponding level to meet the diverse needs of users. The control management of the system adopts the structure of B / S (Browser/Server), including the Web management interface of the foreground and the daemon process of the background. It realizes the remote visual management of the array, and effectively solves the monitoring, processing and recovery of the dual computer system faults. The test results show that the system meets the design requirements.
【学位授予单位】：华中科技大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：TP333.35
[Abstract]:As the core component of the array, the performance of the array controller has become the key factor to affect the overall performance of the array system. The corresponding single point of the disk array storage system starts to transfer to the controller side gradually. The redundancy mechanism of the controller ensures that the system will not be paralyzed and can still provide continuous service in the case of single point failure at the controller end. However, this approach only avoids a single point of failure in the application, and today there are higher requirements for data security, especially in some departments that need to provide uninterrupted service of 724, and disk array systems not only need redundancy on the control side, The reliability of the data store should also be fully guaranteed. Dual controller disk array based on DRBD (Distributed Replicated Block Device) is a dual computer and double cabinet solution, which provides network storage service based on iSCSI (Internet Small Computer System Interface). The storage virtualization of device level is realized through logical volume management, which makes the management of disk devices more flexible. In addition to using two controllers to guarantee the utilization of the equipment and avoid the single point of failure of the controller, the data mirror function of DRBD is used to realize the redundancy of the stored data. The dual fault-tolerant function of controller and storage device is realized, and the availability of the system is improved. By dividing storage devices into ordinary volumes and high availability volumes, the system provides storage services at the corresponding level to meet the diverse needs of users. The control management of the system adopts the structure of B / S (Browser/Server), including the Web management interface of the foreground and the daemon process of the background. It realizes the remote visual management of the array, and effectively solves the monitoring, processing and recovery of the dual computer system faults. The test results show that the system meets the design requirements.
【学位授予单位】：华中科技大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：TP333.35
【参考文献】
相关期刊论文 前10条
1 巩天宁;周书明;;基于DRBD的Linux高可用集群[J];电脑与信息技术;2012年01期
2 唐忠;;双机容错服务器解决方案的研究[J];大众科技;2008年06期
3 申志冰,罗宇;利用Heartbeat实现Linux上的双机热备份系统[J];计算机工程与应用;2002年19期
4 刘军,崔宝江,谢广军,刘t
本文编号：2382034
本文链接：https://www.wllwen.com/kejilunwen/jisuanjikexuelunwen/2382034.html
