How to protect the business continuity of the cloud

In recent years, cloud computing has become a hot topic in the IT world, with the idea of providing IT resources as a service, with goals consistent with IT development: lower costs, faster deliveries, and more diverse services. Cloud computing is not a specific technology, but a service concept that enables the ease and automation of IT services through an on-demand elasticity of expansion and user self-service.

The various categories of the Count cloud

Cloud computing can not be separated from the underlying infrastructure construction, cloud computing technology: currently mainly server virtualization (Vmware,hyperv,xen, etc.), storage virtualization (Fei Kang, IBM, HDS, etc.), network Virtualization (CISCO), Desktop virtualization (Citrix, Oracle-sun).

From the deployment approach, cloud computing contains three kinds: public cloud, private cloud, mixed cloud. For an enterprise, the first consideration is the private cloud approach, which is a good choice for companies dealing with data protection and service-level issues. A dedicated cloud is an on-demand infrastructure owned by a single customer that controls which applications run. They have servers, networks, and disks, and can decide which users are allowed to use the infrastructure.

In terms of delivery, cloud computing is divided into three delivery ways: Software as a Service (SaaS), the best known example of SaaS is Salesforce.com, platform as a service (PaaS), and business examples include Google App Engine, The most common is Google Maps and Google search, infrastructure as a service (IaaS), and the most famous business example is the Amazon Web Service (AWS).

The cloud ERA requires more advanced business continuity and data protection

The advantage of cloud computing is the optimization of resources and the convenience of IT services, and how to ensure the continuity of business operations and the security of core data has always been a priority in reducing it costs. On the premise of universal application of virtualization, how to combine the mainstream virtualization technology to protect the system and data in the existing virtualization architecture. Many users have used VMware for server virtualization and the implementation of the unified storage of data, people familiar with VMware technology know that VMware virtual machines are generally stored in the form of files in the background storage, in this case how to do the corresponding contact data protection? When a virtual machine fails, How do I recover quickly from a background disk array failure? To ensure continuity of the business, has become a new generation of data centers facing problems, Fei Kang Advanced Continuous Data protection technology is adapted to the above requirements, not only provides the core data protection, but also can protect the scope of the operating system to cover the protection of the layer This is the perfect combination of virtualization computing. The continuous data protection technology is not only a disaster backup technology that protects the latest data when the cloud platform disaster occurs, but also a backup technology that restores the original data when the historical data is lost, because the data is saved at multiple points of time automatically, so even if the lost data has been passed for several days, You can still easily achieve the ability to get the right historical data in 1 minutes. For the cloud environment, host backup is really server-free backup way.

From a business deployment approach, a business can choose to use a service provider's cloud or build its own private cloud. Many large enterprises understand the economic benefits of cloud computing, but want to ensure that security policies are strictly enforced. Therefore, the first consideration is the "private cloud", which can increase the economic benefits of the resource pool and standardize its development, deployment, and management. It is also a matter for enterprises to consider how to communicate with each other "private cloud" in time, and how to efficiently complete offsite backup and disaster-tolerance of data between private cloud.

A complete cloud data platform for disaster tolerance, should not only consider switching (failover-from one cloud to another), you should also consider the failback (failback) problem. In general, the cloud Platform production environment in disaster recovery needs to adopt a failback approach, and failback involves more technical points than is switching, so the current market most of the disaster preparedness technology does not have the ability to failback. The company's CDP continuous data protection technology is different, its replication technology provided by the "reverse" power can easily achieve failback after the failure, the data of the disaster preparation cloud synchronization back to production cloud environment, and the drift between the cloud only need to click on the console "reverse" can be completed, simple and convenient.

In addition, in order to ensure the data consistency between cloud production and disaster preparation cloud in cloud environment, the replication technology of Fei Kang Company also provides the consistency contrast mechanism of disk system at both ends of the cloud platform, which can be used to compare the disk block level between the production cloud and the disaster preparation cloud. This technology can be in the production cloud environment and disaster preparedness cloud environment data inconsistency with the rapid completion of Datum point synchronization, the scan difference technology is used to ensure the consistency of data between different cloud platforms, and only when the data between production cloud and disaster preparation cloud is identical can the free drift between different clouds be ensured.

For cloud platforms, replication technologies between different cloud platforms should have a variety of options and adjustments. Let's take a look at what the company has to deal with cloud data drift in different cloud environments: 1. Replication cdr;2 based on continuous IO. A specific time of day. such as: At 12:00 every night to start copying; 3. The duration interval. such as: Copy every 10 minutes, 4. According to the volume change. For example, when new data exceeds 5MB, replication begins. These strategies can be used individually or in combination, providing a very flexible policy triggering mechanism for administrators.

With the continuous expansion of the cloud platform, the transition from private cloud to public cloud is an inevitable trend of cloud platform development, but the transmission bandwidth occupied by data movement between cloud platforms is also a problem that has to be considered and faced. Also mentioned here are Fei Kang's microscan simple data transmission technology. This technology allows data to replicate between different cloud platforms through a variety of connection links, such as SDH, ADSL, T1, T3, E1, E3, F.R, FC, etc. Unlimited distance between local and remote cloud storage pools, can be spanned, across cities, across provinces, and across regions, and the link bandwidth can be flexibly selected according to the increment of data, the input cost can be measured and controlled, and its transmission bandwidth is only the 1/10-1/20 of traditional storage bandwidth, and the cost of investment is minimized. Get the maximum ROI.

Make disaster recovery automation technology a standard service for cloud architecture

The goal of cloud computing is to implement self-service IT services, emphasizing the ease of IT services, and how to ensure continuity of the business is also a problem for a new generation of data centers in the event of an unavoidable disaster or failure (fire, earthquake, physical failure, logical failure), and therefore for disaster, The automation of failback is also an important issue to consider in cloud computing, and based on these issues, the recently launched Recovertrac Technology of the company aims to reduce the complexity of disaster recovery by automating the tasks associated with disaster recovery. Focus on the four "perspectives" of the Data center:

Component perspectives (such as blocks, files, LUNs, objects, transactions)

System Perspective (for example, a single server that supports a single application)

Applying perspectives (such as systems, servers, and combinations of application-related tasks)

Service level Perspective (for example, all applications that include a set of services and have a dependency on cognitive relationships)

Recovertrac has established four levels of perspective, and users can determine the perspective they need. The software automates the many processes of the disaster recovery planning and execution phase, enabling users to locate different components that contain services, and then plan and define the process of disaster recovery. "Users can define recovery policies and jobs for the entire site or for specific services."

Recovertrac also allows IT organizations to test their disaster recovery plans without disrupting major workloads. In addition, the software automatically completes the failover (to Dr site) and failback (to the primary datacenter) operation.

Recovertrac is run in a virtual, physical, or virtual/physical hybrid environment and is compatible with virtualization platforms from VMware and Microsoft. The software can be used for physical to physical, physical to virtual and virtual to virtual recovery.

How to solve the cloud platform data protection problem of X86 architecture?

Cloud computing platform is widely used X86 architecture of PC Server and Windows operating system, and Windows based on patches, system vulnerabilities, and other reasons, is not a very stable system, so how to protect the data while protecting the system is also one of the challenges of cloud computing, This unified protection and recovery is often referred to as "bare metal recovery", flying Kang CDP technology is the best way to achieve this goal at present, the main advantage is: 1. Bypass mode, no change in the original IT architecture, no risk of deployment 2. A solution can protect the operating system and core data 3 uniformly. You can restore the operating system and data to any historical point in time, and so on. These features are adapted to the new generation of data center construction security requirements, better safeguard the continuity of the business.

In the architecture of cloud computing, many enterprises have built their own "private cloud", in the data security and disaster recovery to follow the national and industry regulations, so how to achieve local data backup and recovery, rapid system recovery, in the "private cloud" between the remote disaster disaster, is a problem to consider when building a new generation of data centers, believe that each enterprise CIO will recognize that the data is the core of all enterprises, enterprises to deal with the business, so that products and services in the future into a competitive advantage, to a large extent, rely on data information, this is the rapid development of IT technology today, has been more proof. Protecting data is therefore critical to the enterprise. At the same time, business continuity also requires enterprise IT operations to prevent all kinds of unknown risks, for most enterprises, to build advanced data backup, System recovery system to become a "cloud security" to consider the focus.

At present, the more effective cloud environment data security solutions are not many, the reason is that the spread of cloud computing environment has led to this situation, but there are a number of outstanding, such as the Flying Kang CDP technology for the "cloud" mode of data to provide continuous data protection, rapid recovery, backup and disaster-tolerant solutions, is one of the representatives. Its solution proposes several new data protection ideas:

Protection of the cloud data center by continuous protection technology: Provides robust protection throughout the data lifecycle, not only protecting data, but protecting the system;

The bypass design of continuous data protection technology is characterized by no impact on cloud computing infrastructure and no change in the original system architecture.

Comprehensive disaster protection, to prevent all kinds of logical and physical disasters, to achieve a high level of local cloud environment recovery;

Open architecture, open choice for cloud storage infrastructure

Accurate data recovery can be restored to any IO technology;

The host under the cloud Platform achieves the server-free continuous backup way;

A variety of replication strategies (Continuous replication mode, compression mode, based on time increment mode, based on the data increment mode and other policies can be used alone or combined);

Easy to achieve from the private cloud, the public cloud environment of local continuous data protection and different geographical cloud platform environment of cloud drift.

In combination, the user-self-service services that cloud computing advocates, as well as the ease and automation of IT services, while data security and business continuity are the foundation of cloud computing, with the use of the company's continued protection of products and technologies, customers can take advantage of the commercial advantages of the cloud environment and enhance confidence in the competition Increase the flexibility of the business while ensuring the stability and security of it architecture.

Although cloud computing has its advantages, the current cloud data protection is not a fully fledged domain, and the cloud data protection of the Pegasus CDP is controlled before the cloud infrastructure is built to avoid the loss to users. The popularization of cloud can not be separated from data security, ensuring system and data security is bound to shoulder heavy responsibilities.

(Responsible editor: Lu Guang)