Shulou(Shulou.com)06/01 Report--

One of the most significant changes in 2014 is that the Shuanghuo data center has gone beyond the stage of conceptual hype and has been accepted by many enterprise users. HDS proposes a new dual-active mode-GAD (Global-Active Device), which realizes dual-active based on storage and provides a solid guarantee for business continuity.

Today, ensuring 7 × 24-hour business continuity and data access anytime and anywhere is no longer the "patent" of large enterprise customers, but the most basic demand of ordinary business users and individual consumers. Accordingly, from simple data backup to data disaster recovery, from the disaster recovery scheme of main-standby mode to "dual-active" data center, IT manufacturers and users have adopted a variety of methods to ensure business continuity. One of the most significant changes in 2014 is that the Shuanghuo data center has gone beyond the stage of conceptual hype and has been accepted by many enterprise users.

The dual active architecture is feasible.

The credit card center of China Everbright Bank has deployed the first dual-active front system of the domestic banking system. There has not been a business interruption in the system since it was launched. Both customers and bank leaders have higher and higher requirements for the stability and availability of the banking system. The dual-active front system can reduce the pressure of load processing and ensure the stability and reliability of the system. After many technical arguments, Everbright Bank believes that it is feasible to build a dual-active front system.

The active-standby disaster recovery mode used to be a mainstream mode to achieve high availability of the system. Mainstream storage vendors also provide solutions that support this model. In the master-standby mode, the disaster recovery center starts and takes over the business only when there is a problem in the production center. Because they are not sure about the technology and system, and have not done a rigorous disaster recovery exercise, in the active-standby mode, even if there are problems in the production system, many enterprise managers do not dare to issue the order of system switching easily.

When Jiangxi surveying and Mapping Geographic Information Bureau launched the new cloud public service platform, it also weighed which kind of data center architecture to adopt: if it still follows the main-standby construction idea, then nearly 50% of the data center resources will be in a cold standby state, and daily access services will not be provided, which will result in a great waste of data center resources. If we follow the construction idea of cloud data center and adopt double-active architecture, we can realize the pooling of data center resources, allow business applications and services to drift online, and ensure continuous operation for 7 × 24 hours. Jiangxi surveying and Mapping Geographic Information Bureau finally adopted the cloud architecture double-live data center model.

In the dual-active mode, there is no master-standby relationship between the two data centers, but they are equal and bear the important task of business operation at the same time. If one of the data centers fails, the business will automatically switch to the other. The double-live architecture design ensures the high reliability and scalability of the system. On the one hand, the dual-active architecture can ensure that the two centers provide access services at the same time, and the utilization of data center resources is maximized compared with the traditional master-standby architecture; on the other hand, the dual-active data center architecture makes the underlying IT infrastructure completely transparent to the front-end business applications, and users can schedule IT resources as easily and conveniently as water and electricity. The dual active architecture makes it easy and easy for cloud applications to land.

The advantage of double living is inborn.

In the traditional master-standby disaster recovery solution, the RTO index is usually more than 10 minutes, it has to be switched manually in the event of an accident, technicians have to undertake a lot of script maintenance work, and the resource utilization of the disaster recovery center is low. The dual-active architecture is different. The RTO is usually less than 1 minute. When a failure occurs, the disaster recovery center can be switched automatically. The disaster recovery center is always available, and the resources of the two data centers can be fully utilized.

RTO is close to zero, improve the resource utilization of the data center, in the event of failure to achieve automatic switching is the pursuit of users, but also the value of dual-live data center. HDS believes that the dual-live data center is an ideal solution to protect enterprise business continuity. The value of Shuanghuo data center is mainly reflected in two aspects: first, improve the level of IT services, business is always online, RTO and RPO are infinitely close to 0; second, reduce the operation and maintenance cost and risk of the system.

HDS proposes a new dual-active mode-GAD (Global-Active Device, which implements dual-active based on storage). The HDS GAD scheme consists of two high-end storage HDS Virtual Storage Platform G1000 (VSP G1000). User data can be replicated synchronously in two storage sets in real time, and external data access services can be provided. The user host can access the data stored in any one. If one storage device fails, the application can access the other storage device without interruption. HDS GAD solution can solve two difficult problems for customers: when there is a problem with the production system, the business can automatically switch to another storage device without downtime, and the distance between the two data centers is within 100km; in HDS's dual-activity scheme, the storage on both sides can be used, all hosts can be used, resources can be fully utilized, and the equipment can be switched automatically to ensure the continuous operation of the business.

HDS GAD can easily achieve dual active storage and load balancing, thus improving the capacity and efficiency of the storage system; when any node fails, the system can automatically switch over without affecting the business; Imaco local read and write can achieve better application performance; simple architecture, few fault points, no need for additional software or tools.

The HDS GAD dual-active solution can also make a smooth transition to the 3DC (three data centers) scheme. In the 3DC scheme, the three data centers have equal status and can work together to provide services for business access in parallel, which realizes the full use of resources and avoids one or two backup centers from being idle, resulting in a waste of resources and investment. In the case of a failure or disaster in one data center, other data centers can operate normally and take over the key business or all the business, so as to achieve the effect of mutual backup, so as to achieve the "fault awareness" of users.

SVOS is the core.

Storage Virtualization Operating System (SVOS) adopted by the HDS VSP G1000 provides the core power of a dual-live data center, which simplifies the design and operation of distributed systems. SVOS supports read / write copies of the same data in two places at the same time, and its dual live design enables cross-mirrored storage between two HDS VSP G1000 systems.

These two HDS VSP G1000 systems are able to accept read / write Icano operations at both ends and be updated continuously. If the disk control cabinet at one site fails, the control cabinet at another site will automatically take over and continue to accept read / write IWeiO operations. The dual live data center solution ensures that the latest storage volumes are always available and supports production workloads on both systems, maintaining data consistency, while achieving data protection.

In the dual active data center, two HDS VSP G1000s are active (Active), and each storage provides read / write services to the host, achieving real dual activity and improving the efficiency of storage use. Leveraging some of the unique features of the HDS VSP G1000, the dual live data center solution ensures business continuity by ensuring that the latest active storage volumes are available for production applications even in the event of a failure of a virtualized control cabinet. HDS GAD can now provide dual-live clusters locally and in the same city, and its multipath software allows applications to access replicated data from the shortest path for higher performance.

It should be emphasized that the selection of software and hardware equipment is only one aspect in the process of building a double-live data center. Demand analysis, program planning and design, implementation, data migration, system management and optimization are also links that can not be ignored. When users choose a dual-active data center solution provider, it is best to choose those vendors that have both service and implementation capabilities. HDS Global Solutions Services can provide enterprise users with dual-live data center planning, design, installation, implementation, transition, migration, integration, management and optimization services to ensure the sustainable operation of the data center.

There is a certain threshold for the construction of dual-active system. First of all, users should make an assessment of their business needs and make clear the requirements for system availability or stability, such as which businesses must be protected by highly available systems; secondly, users should realize that the use of dual-activity mechanism can save a lot of resources and costs, improve the utilization of equipment, and easily cope with business peaks. Finally, improving the availability of the system should be considered from the overall architecture, including reliable underlying hardware platform, dual-active technical architecture, upper software and so on.

Shanghai Tonglian Financial Services Co., Ltd. (hereinafter referred to as Tonglian Financial) is a third-party financial service provider mainly engaged in bank card business system outsourcing services based on financial information technology development. In order to meet the growing business needs, Tonglian Finance should establish a centralized, easy to manage and upgrade unified storage platform, and finally achieve the goal of storage "cloud". With the help of HDS's VSP and USP V's disaster recovery technology HUR, Tonglian Financial has realized unified disaster recovery between cloud storage and disaster recovery centers in production centers, and achieved two-way replication between the two data centers, forming a dual-live data center architecture. Based on HDS's global dynamic storage virtualization technology, the dual-live solution not only simplifies and automates IT system management, but also has high availability to ensure the sustainable operation of critical data and applications. The dual-active architecture lays the foundation for unified storage, unified management, unified replication, unified mobility and unified disaster recovery.

To promote the application of the dual live data center solution, HDS has worked deeply with strategic partners such as Microsoft, SAP and VMware to ensure that the HDS VSP G1000 is certified in some important projects, such as Microsoft's private cloud deployment project, SAP HANA custom data center solution, VMware software-defined data center, etc.

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.