A cloud is a collection of tiny water droplets on the 1,000 meters or even higher. The clouds are diffuse and changing all the time form. There are actually a lot of similarities between such clouds and the clouds in computing. The digital versions are a group of resources that often are linked via the Internet and which you also can access the road.Scalability and virtualization are two key words.
The basic idea behind cloud computing is that parts of a network’s resources, or even the entire network can be moved around on the net and managed by another company.It may be easy for small businesses that are unable to have their own IT staff and IT equipment to any great extent. It can of course also be a good solution for larger companies who want to cut down on the number of IT administrators. A move to cloud services means that internally not need as much management-intensive hardware and software. Resources can be something as simple as a single software used on personal computers, but it can also involve a large part of computer hardware.Workstations must be maintained to enable users to access the internet, but you can rationalize away such as storage and database servers. Cloud computing is far from being a novelty. Storage Services via the Internet, for example, existed for nearly a decade. Wide range of services
There are lots of simple cloud-based services including word processing and photo editing and the range is constantly growing. A good example of hardware that moving into the cloud is just storage services. Service providers are content but not with it, but working to move the processing power and even the entire internal network on the internet. Cloud computing differs from traditional IT by being scalable pools of resources that can be easily adapted in various ways. Resources in the cloud requires very little management of the companies that buy in services and the services are consumed in a different way than traditional IT. Companies need only purchase just the resources needed, which is rarely the case with conventional equipment which usually have excess capacity which cost unnecessary dollars.
That is how the companies on cloud services
14% have no plans to use cloud services. 64% see the data cloud as part of its strategic investment. 36% believe that it creates a cost savings of using the cloud.64% plan to use internal IT systems and cloud-based services. 71% worried themselves to increased costs associated with a move to cloud services. of 29% are skeptical about the cloud because they believe that safety is not good enough. Source : Avanade. statistics come from a global survey which questioned more than 500 managers and policy makers at companies in 16 countries.
Public and private clouds
The most common computer clouds are public ones found on the internet. There is also a variant that is within a single network. They are called private cloud, internal cloud or corporate cloud and is still perhaps the most interesting for very large companies and organizations. When building private clouds is based on the same basic structures used in public contexts, but the cloud is not part of the internet. The benefit of the private cloud is that you get some of the advantages of a public cloud (particularly the coordination of resources) while offering a higher level of security because the cloud is not as easily accessible. In essence, the bottom is a private cloud a fully virtualized internal environment, which may consist of a mix of external and internal resources.When you mix external and internal cloud is usually called the solution of a hybrid cloud.These resources can be controlled via a centralized interface so that you can quickly distribute computing power to the points where it is needed most.
There are essentially three basic types of cloud computing: public clouds, private clouds and hybrid clouds. Typical Internet services available in the large, public cloud.
Storage for Virtualization
If we follow the research firm Gartner model is server virtualization only one of several necessary layers to create a cloud. Another important aspect is the virtualization layer between applications and distributed computing resources. It fixes you with a so-called meta-operating system, such as Virtual Datacenter Operating System from Vmware. It is needed to manage the resources of the overall computer pool. Gartner also mentions a layer they call a service distributors, which include making decisions about where resources should be retrieved and what applications will be prioritized.
“Private clouds follow
not the original
principle of cloud computing. “
Some argue that private clouds too much going from the basic idea of cloud computing, as companies that use these themselves must provide hardware and software and also manage them. Private clouds simply follow not the original principle of cloud computing and many believe that it should instead sort them in traditional virtualization.Five different layers
To describe the public cloud computing, one can use a model with the five layers servers, infrastructure, platforms, applications and clients (see chart below). Some say that there should also be a sixth store to store. There are also different views on what stocks should be named.
Server layer is a set of cloud custom servers and associated hardware and software, many of which are scattered in different locations and are interconnected with super-fast network connections. Hardware and software can of course also be assembled in one place. The second layer is popularly known as the Infrastructure as a Service, IaaS, and just as it sounds include the infrastructure as a service. Usually this is done by the service provider to offer virtual machines or operating systems. The customer buys the necessary capacity and thus have no need to purchase servers, network equipment or software. The service provider will usually charge for usage of the service and the possible storage space used. Formerly called the IaaS for Haas, an acronym for hardware as a service. Cloudy platform
A cloud platform, platform as a service (PaaS), offers a hosted service for companies looking to develop and test applications without having to invest in special hardware or additional software. It is based on cloud infrastructure and can be utilized in a variety of contexts, such as by a developer who takes out a SaaS service and do not want to build a platform for the purpose. Paas may also include services for team collaboration, integration of web services and other similar variations. Paas may also be a definition of services intended for ordinary users including those that offer a desktop-like environment, a form of operating online. There, the user can, for example, playing games, using everyday applications, and store personal files. Some believe this is the future for most computer users and that it only becomes necessary to have an internet terminal at home. The downside is that you are limited if your Internet connection goes down or if the service provider have technical problems. Applications included in the cloud, software as a service (SaaS), are those that generally can be run directly in a browser without having to install any software locally on the computer. The advantage is that IT administrators do not have to keep up with updates, and usually not have to worry about which platforms the applications supported. Examples of simple SaaS services are common sites like Facebook and the like, but usually the intention is for more advanced services such as Photoshop.com , Microsoft Office Live Workspace or Google Apps. Usually when people talk about cloud computing is SaaS provided. When services are specifically dealing with security issues is called the security as a service.Clients are hard or soft
Clients are final layer in the chain and they can be both hardware and software that is directly related to a cloud service in order to function. For programs Panda Cloud Antivirus is a very good example. It is an antivirus program that needs to connect to a virus database on the network when it is time to perform a virus scan. It does not store any virus definitions locally on an end user’s computer.
“Hardware can be
all sorts of things, such as
advanced cell phones
or thin clients. “
Hardware can be all sorts of things, such as advanced cell phones as the iPhone or models with Android or some type of thin clients that run their applications in the cloud. If you choose to follow this particular model can be used in other words, divide the cloud services in the three main categories IaaS, PaaS and SaaS. Below them are also more as a service variants, such as storage as a service, that confused probably shortened the same way as software as service. Interface for Communications
A cloud is made up of a number of cloud components that communicate with each other over some kind of API (application programming interface), an interface that sets up rules that allow components to talk to each other. Usually it is some type of Web service that handles it. As previously mentioned is the components of both hardware and software and they are coupled together by the service provider so that they can deliver the performance, capacity and features that the end user expects. To get cloud services work is also needed some form of central administration area that holds together the loosely assembled parts. Requires high capacity
Many cloud services require computing power beyond the ordinary and this is where the term fabric computing enters the picture. It was first used at the end of the 1990s, but has recently become more relevant than ever. With fabric computing means a high performance computing (hpc, high performance computing) system consisting of storage space, computing resources and other necessary hardware are connected via connections that offer extremely high bandwidth. connections are such as Infiniband, ten gigabit ethernet, fiber channel or some other variant that provides gigabit speeds. The important thing is that all the computing power is emerging as a single unit.Enormous computing power
They usually refer to the most elementary things in fabric computing nodes, which is simply the components such as memory, CPU and any peripherals. The nodes are connected by links, the second basic element when it comes to fabric computing. The big advantage is that you get a system with a huge collective computing power that can be used for the toughest tasks. term fabric computing is used in some contexts also to describe the general grid computing, and platforms such as Microsoft Windows Azure.The latter is actually a PaaS service that offers a wide range of services that can be used by other services on the Internet or by traditional programs that need to be supplemented by a cloud service. The foundation of the platform is a specialized network operating system that bears the same name as the service.
“In the cloud occurs
MPLS technology to get
higher performance. “
The cloud appears that traffic is routed by MPLS technology (multi-protocol label switching) to improve performance and we are then told that it has created a specific MPLS cloud. It is useful for service providers operating in the example video conferencing sector that transmits large amounts of data and where it is important that communication is robust and always at a high level. Mpls creates virtual links between nodes in a network and ensures that a packet is transported on safely from point A to point B. This is done by groups of packages bear a special mark, which then serves packets only guide. There is no need to analyze the content itself, something that is otherwise customary. This allows each router to make a quicker decision than normal Layer 3 routing. In addition to improved performance is the advantage that it does not rely on any variation of the techniques fall into OSI model Layer 2. Mpls is usually placed in between layer 2 and layer 3 and is therefore called for a layer 2.5 protocol.
A highly simplified picture of how an MPLS cloud can look like. LSR routers handles traffic, while the internal edge LSR routers acts as an interface between MPLS and traffic data are routed in a conventional manner.
Using a table
To routers on a network, understand where a group of MPLS-labeled packet is sent, a table called label forwarding information base, lfib. It is a combination of the usual routing table content and information on special labels look for different destination networks. Each router tells its neighbors which marks it knows and then constructed local lfib tables in the routers. The latter is done via the protocol label distribution protocol, ldp. To the traditional routed packets can be transported in an MPLS network, they must pass a elsr router (edge label switch router) to notice about them. The packets are then sent to internal routers LSR (label switch router), which updates the label with current information so that packages can pass through the MPLS cloud as quickly as possible. At the end of the chain will always packets to another elsr router deletes mpls marking. The packets can then travel on as normal