[an error occurred while processing the directive]
RSS
Логотип
Баннер в шапке 1
Баннер в шапке 2
2015/04/22 15:32:41

Cloud computing

Cloud computing (English Cloud - cloud; computing are calculations) - "cloud computing" - the concept of "a computing cloud" according to which programs are started and give results of work in a window of the standard web browser on the local personal computer, at the same time all applications and their data necessary for work are on the remote server in the Internet.

Computers performing such calculations are called the "computing cloud." At the same time, the load between computers included in the "computing cloud" is automatically distributed. The simplest example of cloud computing is p2p networks.

In the 2010s, cloud computing was a group of technologies that led the development of information technology in general, which has even more influence than electronic business at one time.

See also:

* Cloud computing * SaaS * IaaS * PaaS

Content

Cloud Computing Development

The concept of "cloud computing" originated in 1960, when John McCarthy suggested that someday computer computing would be made using "public utilities."

Cloud computing may seem like a relatively new phenomenon. However, their history is rooted in the early 1950s, when the advent of mainframe systems allowed several users to access the central computer. In the 1960s, some ideas appeared that resemble what we call cloud computing today - for example, the concept of the "intergalactic computer network" of J. K. R. Licklider[1].

The ideology of cloud computing gained popularity in 2007 due to the rapid development of communication channels and the growing exponentially growing need of both business and private users to scale their information systems horizontally.

In the 1970s, virtualization raised mainframe to a new level, and in the 1990s, telecommunications companies began to offer connection to a virtual private network (VPN). In 1999, Salesforce became the first company to provide enterprise applications over the Internet. Several users could download these applications simultaneously in the browser at a low price.

Image:cloud.jpg

Later, the concept of a "computing cloud" began to be actively used by various companies, for example, Google. The most typical example  is Google Docs, which allows you to work with office documents through a browser.

Modern "clouds" appeared in 2006, when Amazon.com, then an online bookstore, introduced Amazon Web Services (AWS), laying the foundation for the movement of cloud computing. AWS provides a wide range of services, such as computing power and data warehouses, to this day remaining the leading and very reliable infrastructure of cloud web services platforms.

Soon, Netflix Microsoft, Google joined the Amazon.com Apple , and, and the IBM cloud computing market has grown.

In July 2008, HP, Intel, and Altaba (formerly Yahoo) announced the creation of a global, multi-site, open Cloud Computing Test Bed for the development of cloud computing research and development. This laboratory was a globally distributed Internet testing environment that supported research aimed at developing software, improving data center management, and solving hardware problems related to Internet computing of a much larger scale than ever before. This initiative was also supposed to contribute to the emergence of new Internet applications and services.

Microsoft did not stand aside: the executive director of the corporation, Steve Ballmer, made a statement that Microsoft plans to release a new operating system codenamed "Windows Cloud," which will allow developers to create and host Internet applications. The name "Windows Cloud" indicates that the new OS will use the concept of a "computing cloud" as its basis.

For 2014, most of the largest IT vendors in the global market, including Google, Microsoft, HP, Intel, SAP, IBM, Oracle and others, have cloud computing solutions in their line.

Cloud Computing Properties and Models

Main properties

NIST in its document 'The NIST Definition of Cloud Computing' defines the following cloud characteristics:

  • the possibility of self-service without the participation of a person by the provider;
  • availability of broadband network access;
  • Focus resources on individual sites for efficient allocation;
  • Fast scalability - Resources can be freed up and allocated at great speed depending on your needs
  • Managed service - The cloud management system automatically monitors and optimizes provisioning based on measured service parameters (storage size, bandwidth, number of active users, etc.).

On-demand self-service. The consumer has the opportunity to access the provided computing resources unilaterally as needed, automatically, without the need to interact with the employees of each service provider.

Broad network access. The provided computing resources are available over the network through standard mechanisms for various platforms, thin and thick customers (mobile phones, tablets, laptops, workstations, etc.).

Resource pooling. The provider's computing resources are pooled to serve many consumers in a multi-tenant model. Pools include various physical and virtual resources that can be dynamically assigned and reassigned according to consumer requests. There is no need for the consumer to know the exact location of the resources, but you can specify their location at a higher level of abstraction (for example, country, region or data center). Examples of such resources include storage systems, computing power, memory, and network bandwidth.

Rapid elasticity. Resources can be elastically allocated and released, in some cases automatically, for rapid scaling commensurate with demand. For the consumer, the possibilities for providing resources are seen as unlimited, that is, they can be assigned in any number and at any time.

Measured service. Cloud systems automatically manage and optimize resources using measurement tools implemented at the abstraction level for various types of services (for example, managing external memory, processing, bandwidth, or active user sessions). The resources used can be monitored and monitored to ensure transparency for both the provider and the customer using the service.

Cloud Service Models

Software as a service (SaaS). The ability to provide the consumer with the use of provider applications that work in the cloud infrastructure. Applications are available from various client devices or through thin client interfaces such as a web browser (such as web mail) or program interfaces. The consumer does not manage the underlying cloud infrastructure, including networks, servers, operating systems, storage systems, and even individual application settings, except for some custom application configuration settings.

Platform as a service (PaaS). Enable the consumer to deploy consumer (created or purchased) applications to the cloud infrastructure through programming languages, libraries, services, and service provider-supported tools. The consumer does not manage the underlying cloud infrastructure, including networks, servers, operating systems, and storage systems, but has control over deployed applications and possibly some configuration settings for the hosting environment.

Infrastructure as a service (IaaS). The ability to provide the consumer with processing, storage, networking, and other fundamental computing resources to deploy and run random software that can include operating systems and applications. The consumer does not manage the underlying cloud infrastructure, but has control over operating systems, storage systems, deployed applications, and possibly limited control over the selection of network components (for example, a host with network displays).

Expansion models

Private cloud. Cloud infrastructure designed for exclusive use by a single, multi-consumer organization (e.g., business units). Such a cloud can be owned, managed and maintained by the organization itself, by a third party and located both on the territory of the enterprise and outside it.

Community cloud and community cloud. Cloud infrastructure designed for exclusive use by a specific consumer community from organizations with common problems (e.g., missions, security requirements, policies). The cloud can be owned, managed and maintained by one or more organizations in the community, and a third party can be located both inside and outside the organizations.

Public cloud. Cloud infrastructure prepared for open use by the general public. It may be owned, managed and maintained by business, scientific and governmental organizations in any combination thereof. The cloud exists on the territory of the cloud provider.

Hybrid cloud. Cloud infrastructure is a composition of two or more different cloud infrastructures (private, public, or public) that have unique objects, but are interconnected by standardized or proprietary technologies that allow data or applications to be transferred between components (for example, for load balancing between clouds).

The Benefits of Cloud Computing

  • reduced requirements for PC computing power (only Internet access is a prerequisite);
  • fault tolerance;
  • safety;
  • high data processing speed;
  • Lower hardware, software, maintenance, and power costs
  • Save disk space (both data and programs are stored on the Internet).

Cloud Computing Flaws

  • dependence of user data preservation on companies providing cloud computing service;
  • the emergence of new ("cloud") monopolists.

Safety

Confidentiality should be maintained throughout the chain, including the cloud solution provider, the consumer, and the communications that connect them.

The supplier's task is to ensure both physical and programmatic inviolability of data from third party attacks. It is no coincidence that cloud data centers are usually designed based on the most modern security standards (including encryption issues, as well as the mentioned anti-virus protection and protection against hacker attacks).[2]

The consumer shall put into effect "in his territory" appropriate policies and procedures that exclude the transfer of rights of access to information to third parties. In this sense, the objective advantages of the "clouds" should not be mixed with getting the customer rid of any efforts to ensure the security of their own information perimeter.

Security solutions include traditional and well-known solutions, although they contain a number of specific solutions that, in the process of performing traditional tasks, must be optimized to save the performance of the virtual environment, adding security.

Even large cloud service providers are already experiencing major hardware failures. In the global practice of cloud computing, there are cases when a consumer could not access applications for a long time. And the banal "Internet shutdown" due to the fault of the provider (not necessarily the provider directly serving the customer, the backbone operator may be to blame) may make working with "cloud" resources impossible in principle.

Obviously, before starting projects related to taking certain IT services to the clouds, customers should assess such risks, conduct a thorough inventory of applications (recording a list of critical for business), and only then make decisions about how to build their cloud IT future.

An alternative ISP that is hot-standby, an alternative provider of a cloud solution, transparent management of maintaining archived copies of data, insurance, strict liability conditions in agreements with suppliers - mandatory security elements in the cloud.

The challenges of ensuring the integrity of information in the case of individual "cloud" applications can be solved - thanks to modern database architectures, backup systems, integrity check algorithms and other industrial solutions. But there is more to come. New problems may arise when it comes to integrating multiple cloud applications from different vendors.

For those companies whose information protection issues are very acute, for example, these are enterprises related to the military-industrial complex, work with state secrets, or rigidly connected by hand requirements for non-disclosure of data about private customers, the way out of the situation is the creation of private clouds. The fact is that private clouds, unlike public or hybrid systems, are most like virtualized infrastructures that IT departments of large corporations have already learned to implement and over which they can maintain complete control. The shortcomings of information protection in public cloud systems pose a serious problem. Most hacking incidents occur in public clouds.

Virtualization can be secure and compliant with information protection regulations. However, customer security requests are still often ahead of vendor capabilities.

Cloud Usage Guidelines

Gartner experts are of the opinion that this particular direction should change the established status quo in information technology in the near future. Cloud computing is expected to spur even more intense Internet development. Gartner predicts that the trend will be finally formed over the next few years.

Analysts also note that cloud computing technology will reduce costs and increase demand for new IT products, but the effect of growth from such technologies will appear only in the long term.

Choosing a cloud vendor is almost like getting married. Both sides are determined for the best and are confident that the relationship will be long and full of love and mutual understanding. But … visibility is deceptive, and relations with the vendor can deteriorate. In such a case, a "marriage" contract is needed (more precisely, a clear and exhaustive contract), ensuring that both parties involved know their rights and obligations.

CRN turned to the leaders of the VAR company Progress Software to gain first-hand knowledge. Matt Chichchari, marketing manager for OpenEdge/SaaS platforms and cloud implementation, and Mike Ormerod, architect of cloud and SaaS solutions, shared their experience: what are the main elements that must be present in the "marriage" contract with the cloud vendor so that the potential client can safely say yes.

Prepare network

During the 2012 Interop exhibition in Las Vegas, Cisco Systems released the results of the annual Global Cloud Networking Survey, which brought together 1,300 IT professionals from 13 countries. Over a third of IT professionals consider a cloud-based network to be an essential part of the infrastructure to begin migrating applications to the cloud. 28% of them believe that a cloud-optimized network is more important than a virtualized data center, 21% - than a service level agreement with a cloud service provider.

At the same time, almost 40% of respondents will prefer in every way to avoid any network restructuring associated with the introduction of private or public clouds. In other words, many IT managers are aware of the importance of building a specialized network infrastructure for cloud computing, but few are willing to implement objectively complex projects.

Praveen Akkiraju, Vice President and General Manager of Cisco Services Routing Technology Group, warns that company leaders should have a clear understanding of all necessary steps before the full-scale deployment of cloud infrastructure. Cisco representatives believe that in the foreseeable future, the number of such projects will grow at an increasing pace. According to the Global Cloud Index report, by 2014, more than half of data center capacity will be spent on cloud computing maintenance, by 2015 cloud traffic will grow by 12 times and reach 1.6 zettabytes per year. 73% of participants have all the necessary information to implement private and public clouds, they consider well-established processes and proper planning as the cornerstone of a successful project.

First understand the essence of the cloud

Progress Software executives warn that the decision to implement the cloud can be tied in terms, acronyms, and symbols. It is important that the potential customer complete the "homework" and get used to the terms that vendors operate before entering the path of negotiations. "You should not immediately throw yourself into a new case if you do not understand some terms and technologies," says Chichchari. "First of all, you need to understand what you are entering into," Ormerod echoes him.

Know all SLA parts

The recent series of cloud failures has taught us something. A clear service level agreement (SLA) is needed to help this union be long and happy. Progress Software executives recommend that customers of cloud services carefully examine all SLA conditions to know who is responsible for what in different circumstances. It is important to ensure that critical applications do not change later; A signed SLA protects not only your reputation, but also the vendor's name. "The text and wording of the SLA is everything," says Ormerod. "As in any contract, you must have absolute clarity on who does what, when and how," adds Chichchari.

Have a work plan

Remember, cloud vendors are not IT management service providers. When you enter the cloud, you have the same responsibilities as using the infrastructure on the ground. The cloud does not belong to the category "turned on and forgot," and here the principle "out of sight - out of heart" does not pass. It is important to plan how day-to-day operations are performed, who has access to what and when, and all aspects of IT security. You also need to understand the escort procedures, whether you eliminate bugs or upgrades. "If you do not see something, this does not mean that you are not responsible for it," says Chichchari.

Have a disaster recovery plan

The union of two is a constant work, and something can go wrong. It is important to plan your actions in case of unexpected problems. Progress Software is ready for this. Before you dip into the cloud, you must have a ready-made plan to recover from failures and recreate the working environment. A high level of readiness is the work of the cloud vendor, but overcoming failures is not, Chichchari warns. "There is such a misconception: I send all my applications to the cloud - and no worries," Ormerod echoes him.

Know where your data is

If the union broke up, then the division of property is inevitable. It's about the same cloud situation. The vendor's contract must specify in detail what happens to the customer's data if he or the vendor goes out of business, if there is a merger or purchase of one of the parties, and how long the vendor must keep the customer's data. Data location and compliance are also important aspects of the transition to the cloud, Ormerod says.

Mnogoplatformny overcast

This may look like "infidelity," but any vendor cloud contract should contain a clause to support various types of cloud with the ability to use other platforms. Vendors should be asked if they provide support for the public, private cloud and hybrid model, says Chichchari. For a relationship to be long, the customer must be able to use multiple cloud vendors simultaneously for the same application, system, or environment. "Why limit yourself to just one cloud?" he says. "What if something happens to Amazon, GoGrid, or Rackspace?" Customers should ask themselves: "How easy will it be for me to transfer my applications from one vendor to another?" - adds Ormerod.

Need a exit strategy

And finally, what if the relationship has not developed, and both parties are forced to part? When the honeymoon has passed, users should know what to do, says Chichchari. What if there is a problem, how do I recover data and restart the cloud application? And if the customer just changed his mind and no longer wants to use the cloud? If your business profile or market has changed and you need to change your strategy? The cloud service contract should provide for a exit strategy so that both the vendor and the customer can amicably part, avoiding unpleasant showdowns.

Cloud Computing: Myths and Misconceptions

"Cloud" is based on software only

Theoretically, it is quite possible to build a cloud on standard servers (x86) and intelligent software. We combine several virtual devices and get a cloud. But in fact, this is far from true. For various reasons, such as maintaining adequate performance (specialized ASICs or dedicated hardware resources), ensuring compatibility (installing drivers for each new x86 platform), or monitoring functions (HIPPA, PCI-DSS, departmental isolation, etc.), not all system developers have yet abandoned the use of dedicated hardware resources for certain elements of their data centers. In principle, the inevitability of virtualization of some components of the computer environment is obvious. Therefore, market leaders produce appropriate equipment. For example, Nexus 1000v, which provides transparency of virtual machine traffic at the network security level, with built-in VN-Link technology that provides network mobility. As well as a virtual security gateway and vWAAS. In some cases, clients choose virtual devices. In other cases, they prefer a combination of software and hardware resources, for example, checkpoints represented by Nexus 1010v. All of these unified network services provide system developers with a standard set of techniques to share software and hardware resources when needed.

Cloud and Elastic Resources

For many network equipment vendors, the idea of ​ ​ elastic resources is not fully realized. The fact is that the concept of "elastic resources" should not be limited only to servers and storage systems, it should apply to all "cloud," including its network elements.

Cloud and Pooled Resources

This is another area that many connect only to servers and storage systems. For many years, before the cloud appeared, the network was a complex resource that provided services (bandwidth, security, segmentation or isolation, QoS function, etc.) to individual groups of systems. As networks became more virtual, more automated, and the number of tenants grew, companies offering solutions in this market continued to expand the intelligent logic of their products required for the functioning of modern cloud systems. These solutions allow customers to deploy modular virtual systems (such as Cisco's Vblock and SMT).

Virtualization makes the cloud more flexible

On the one hand, this is the absolute truth with respect to the dynamic allocation of resources and the rapid inclusion of new virtual machines using templates and clones. However, this requires some structural changes. For example, if you are using vMotion, Live Migration, or XenMotion, the source and endpoint must be in the same logical domain. And surveys of virtual system administrators show that in a month they have to perform thousands of virtual machine migrations. So, in addition to this flexibility, they only have to abolish the traditional distinction of levels. To do this, they need to be given the opportunity to build large, two-level networks, devoid of constant problems of hierarchical networks. And it would be even better if this concept covered not one, but many networks, thereby greatly simplifying the problem of application mobility and accident prevention. Just as the network began to support the transmission of various multimedia data in real time, it must now adapt to the changes associated with virtualization and dynamic computer environments.

A virtual machine is the same server, software only

That's not exactly true. Indeed, the application and the guest operating system probably do not see any difference, but network administrators and the security service do not care at all. The virtual machine no longer starts or ends at one end of the Ethernet cable. In fact, its traffic may not be related to this cable at all if it communicates with another virtual machine within the same host. Or it can switch from one host to another several times a day, thereby occupying gigabytes of traffic (virtual machine and storage system). Therefore, it is much more prudent to introduce the registration function of all these migrations in the cloud, assign certain policies (security, QoS function, role-based access system, etc.) to these virtual machines to limit their movement to certain places, and thereby expand the capabilities of the technical support service. Most virtual machine administrators would be happy to have such capabilities, especially now that they are demanding data on virtual machines, backup, disaster recovery, guest operating system, storage, and all other elements that are somehow related to virtualization.

Clouds are planned and networks are administered

It is possible that it is, but it implies that your computer environment is still isolated from your network. You can look at things in a different way. Technologies such as virtualization require not only closer integration of computing and network operations, but also a higher level of automation and control, a level of automation that provides open access to all stand-alone systems and functional modules with specified policies.

Impact estimates for EU economies

In the December 2010 report "Cloud Dividends - 2011," the Center for Economic and Business Research (CEBR) claims that by 2015, thanks to cloud computing, the economies of developed European countries will receive an additional 177.3 billion euros per year. The EMC-commissioned report was the first-of-its-kind assessment of the macroeconomic value of cloud computing for Europe's top five economies.

The authors of the CEBR report concluded that if in the UK, Germany, Italy, Spain and France the introduction of cloud technologies will continue at the expected pace, then by 2015 they will bring the economies of these countries at 177.3 billion euros per year. It is important to note that the lion's share of these funds, as the study shows, will be provided through the development of private and hybrid cloud computing models.

CEBR estimated that the annual economic impact of cloud computing for each country by 2015 would be: Effect, billion euros:

  • Germany - 49.6
  • France - 37.4
  • Italy - 35.1
  • United Kingdom - 30.0
  • Spain - 25.2

177.3 billion could cover loans made to some of the region's debtor countries, such as Ireland (85 billion euros) and Greece (110 billion euros), and help the UK government implement a plan to cut public spending by 95.7 billion euros in four years, which it recently announced.

Cloud computing is a new approach to IT, in which technology becomes available to enterprises at the right volume and when they need it, the study says. This accelerates time to market, removes traditional entry barriers, and enables companies to leverage new business opportunities. By increasing competition, this direct effect of cloud computing will have a huge impact on the market structure in many sectors of the economy, and therefore on global macroeconomic indicators, says CEBR.

CEBR believes that cloud computing will be an important factor in economic growth, competitiveness and the creation of new enterprises throughout the eurozone. This underscores the importance of this technology for the economic recovery of the region, particularly in the face of the growing threat posed by emerging economies, which traditionally benefit from more intense competition.

The study focuses on three of the most common cloud computing models:

  • A public cloud that is under the control of the service provider
  • private cloud under the control of the organization's own IT department;
  • hybrid cloud, which is a combination of the first two models.

CEBR predicts that by 2015, 133 billion euros, or 75% of the total annual economic effect of 177.3 billion euros, will fall on non-public cloud computing models. The private cloud model allows you to kill two hares at once: organizations receive dynamic, on-demand, self-service, and scalable cloud computing services, but control remains in the hands of the IT department, so security and manageability requirements are not violated.

During the study, CEBR also found that the private cloud will contribute to accelerating the pace of development and creation of new enterprises in the amount of 23.8 billion euros. The resulting indirect and derivative investments and total expenditures will create additional demand for goods and services, which in turn will increase gross value added (GVA) and employment in the economy. The CEBR predicts that by 2015, the indirect economic benefits from the additional GVA in all five countries will total 280 billion euros - at 60 billion per year - and that indirect and induced employment between 2010 and 2015 could reach 2,396,000 workers.

CEBR lead economist Oliver Hogan noted: "The CEBR study shows that cloud computing is not just a matter of short-term improvement in the efficiency of IT investments of individual companies and, therefore, their productivity. This technology could be a critical macroeconomic factor that will be critical to stimulating economic growth in Europe, especially in today's uncertain economic situation. As a factor in improving performance, cloud computing is likely to play a particularly important role in ensuring that Europe remains competitive in the global market, and therefore increases export growth. Moreover, as one of the main modern tools to achieve maximum efficiency of investments in IT, cloud computing can also become the locomotive of European business investments, which, in turn, will move European countries forward. "

EMC President of EMEA Rainer Erlat believes: "The mobility and competitiveness that private and hybrid cloud computing gives enterprises create real opportunities for European businesses - they will help companies increase their advantage by contributing to the economic recovery of their own countries. It is generally recognized that achieving economic recovery and maintaining economic stability requires debt reduction while promoting commercial competition. Cloud computing, which will replace many of today's IT technologies by offering more efficient, flexible, and simple solutions, is the real way to do it. "

Cloud computing will help companies not only take advantage of business opportunities, but also achieve significant cost savings. The payment model for services actually received leads to a decrease in capital costs (CAPEX) and current costs (OPEX), a quick return on investment and a more efficient reallocation of resources. These savings can be reinvested by encouraging innovation, increasing competitiveness and directly improving profitability, that is, having a tangible positive effect on the economies of countries.

Research methodology

The Cloud Dividend report calculates the savings (capital costs and operating expenses) that companies receive from the implementation of cloud computing services, and measures the impact of these savings on macro and corporate economic performance, such as favorable business opportunities; establishment of new enterprises; indirect gross value added (GVA); contribution to the payment of taxes; and the cost of cloud computing services to determine the economic value of the technology for each country. Cloud Dividends - 2011 is the first in a cycle of two reports. The second report, which will be released in February 2011, will look at the economic impact and impact of cloud computing on specific sectors of the economy in France, Germany, Italy, Spain and the UK.

The report can be downloaded here

Cloud Model Forecasts

Шаблон:Main'Cloud computing: 10 changes that will occur to them by 2020

We are in the early stages of cloud computing (Forrester, August 2012). Many organizations take only the first, insecure steps. But by 2020, the cloud will become the main - and indispensable - part of the enterprise's computing infrastructure. Further...

Notes

See also

Links