What are different methods of virtualization?

I am just aware about the SAN virtualization. Traditional approaches to integration of advanced features in SAN environments (including blades and boxes), still have to contend with other problems. This allows, for example, chassis-based blade but manage well, but what are their terms of performance and scalability, greatly limited by the heat and power consumption. Now I want to know more methods of virtualization. Please tell me what are different methods of virtualization? I am sure that you members will help me soon (as usual).

Some virtualization architectures are self-contained. That is, the control and translation of the abstract can be found in the same device. Implementations include architectures with self-contained appliances and solutions to the array controller-based. In this approach, the limits imposed by lack of scalability make it particularly noticeable. For solutions that work according to this pattern are only to be able to virtualize certain parts of the fabric. The products are not all the memory it comes to so-called. Virtualization Islands, so virtualized parts of the total space that can not communicate with the other storage areas. Data that reside on such islands, can not move without affecting the overlying applications.

An approach that encompasses the entire storage network, and is penetrates itself better to solve the difficulties connected with the scalability and to meet the business needs. In this approach, the virtualization is not limited to specific devices or parts of the storage network. Therefore, one can with the above procedure, also called the network - is called Fabric-based method or, virtualize the entire SAN. An approach that encompasses the entire storage network, and is penetrates itself better to solve the difficulties connected with the scalability and to meet the business needs. In this approach, the virtualization is not limited to specific devices or parts of the storage network. Therefore, one can with the above procedure, also called the network - is called Fabric-based method or, virtualize the entire SAN. Most virtualization strategies based on fabric to work with a separate data path, while the physical input and output is handled by an independent control path. This technique works similar to a disk subsystem. The control path "programmed" the data path and at the same time, the logical abstraction of physical storage array LUNs for other tools, such as storage provisioning and data management available.

In practice, it looks like this: When starting up a computer takes over the control path, the task of cataloging the existing plates and a mapping table to generate (virtual vs. physical addresses.). The system creates this table, then into an ASIC. If, in operation an application requests data, the request goes first to the virtual address. Specifying a particular port is addressed. Behind this port is operating an intelligent, high-performance processor (ASIC), which converts the request through the mapping table so that the application, the right data from the correct physical memory unit is replaced. All these processes occur via the data path. In this way, the customer data never leaves the fabric. The control path is therefore used only for answering inquiries and execution errors. The volume of data transmitted over the control path control data is a total of only about two percent of the total transfer.

Quote:

---

The control path is therefore used only for answering inquiries and execution errors. The volume of data transmitted over the control path control data is a total of only about two percent of the total transfer.

---

The advantage of this technology lies in the speed: Only storage components with separate paths provide the necessary combination of high performance and scalability. The separation between the data and the control path in the industry in general as Spaid architecture (Split Path Acceleration of independent data streams, including: Split Path Acceleration of Intelligent Devices) called. She makes an independent scalability of virtualization performance for data path and level of application performance on control path level possible.

Provider of powerful virtualization solutions rely on highly scalable single platforms that are capable of supporting a high level of performance many different software applications without the need here to implement it for each application its own hardware platform. In this way they have the option of a single investment for a number of software solutions to use. In future, the flexible storage platforms also support additional fabric services such as software-based virtual tape libraries, replication and continuous data protection. The most innovative products can even integrate seamlessly into existing production environments without having to replace existing components need. It is not necessary in such cases, existing storage components (and investment) to write off or make major reconfigurations. Importantly, however, that ultimately any additional functionality to the entire storage network should be available and not on specific ports or devices remains restricted to prevent the formation of islands. One approach to eliminating the physical memory limits of limits must be no new restrictions on the network to create.

In operation, let's modular solutions using standard Fibre Channel ports to connect to the directors. The directors "absorb" then turn the management of the module and make it the whole fabric as virtualization service. Thus, the above-mentioned products differ significantly from independent switches and appliances, which the administrators have to manage each of its own. The server and storage components of the production network remain unaffected by the whole process. In operation, manages the fabric to as "virtualized" traffic for certain transactions to the storage virtualization engine, and this then performs the necessary line-rate transactions. The resulting benefits are obvious: The virtualization modules have their own cooling systems and power supplies can be adjusted to reflect new technological requirements, and across generations in terms on Fibre Channel devices. In turn this means in practice that existing customer environments, which in the past three years, with Fibre Channel Directors established, can be upgraded to this new fabric services. The same applies to current purchases and the implementation of future directors. Thus, these modules deliver a unique investment protection across multiple Fibre Channel Director generations.