A good indication any new technology or business model is starting to mature is the number of certifications popping up related to that product, framework, or service. Cloud computing is certainly no exception, with vendors such as Microsoft, Google, VMWare, and IBM offering certification training for their own products, as well as organizations such CompTIA and Architura competing for industry neutral certifications.
Is this all hype, or is it an essential part of the emerging cloud computing ecosystem? Can we remember the days when entry level Cisco, Microsoft, or other vendor certifications were almost mocked by industry elitists?
Much like the early Internet days of eEverything, cloud computing is at the point where most have heard the term, few understand the concepts, and marketing folk are exploiting every possible combination of the words to place their products in a favorable, forward leaning light.
So, what if executive management takes a basic course in cloud computing principles, or sales and customer service people take a Cloud 101 course? Is that bad?
Of course not. Cloud computing has the potential of being transformational to business, governments, organization, and even individuals. Business leaders need to understand the potential and impact of what a service-oriented cloud computing infrastructure might mean to their organization, the game-changing potential of integration and interoperability, the freedom of mobility, and the practical execution of basic cloud computing characteristics within their ICT environment.
A certification is not all about getting the test, and certificate. As an instructor for the CompTIA course, I manage classes of 20 or more students ranging from engineers, to network operations center staff, to customer service and sales, to mid-level executives. We’ve yet to encounter an individual who claims they have learned nothing from attending the course, and most leave the course with a very different viewpoint of cloud computing than held prior to the class.
As with most technology driven topics, cloud computing does break into different branches – including technical, operations, and business utility.
The underlying technologies of cloud computing are probably the easiest part of the challenge, as ultimately skills will develop based on time, experience, and operation of cloud-related technologies.
The more difficult challenge is understanding the impact of cloud computing may mean to an organization, both internally as well as on a global scale. No business-related discussion of cloud computing is complete without consideration of service-oriented architectures, enterprise architectures, interoperability, big data, disaster management, and continuity of operations.
Business decisions on data center consolidation, ICT outsourcing, and other aspects of the current technology refresh or financial consideration will be more effective and structured when accompanied by a basic business and high level understanding of cloud computing underlying technologies. As an approach to business transformation, additional complimentary capabilities in enterprise architecture, service-oriented architectures, and IT service management will certainly help senior decision makers best understand the relationship between cloud computing and their organizational planning.
While reading the news, clipping stories, and self-study may help decision makers understand the basic components of cloud computing and other supporting technologies. Taking an introduction cloud computing course, regardless if vendor training or neutral, will give enough background knowledge to at least engage in the conversation. Given the hype surrounding cloud computing, and the potential long term consequences of making an uniformed decision, the investment in cloud computing training must be considered valuable at all levels of the organization, from technical to senior management.
The current technology refresh cycle presents many opportunities, and challenges to both organizations and governments. The potential of service-oriented architectures, interoperability, collaboration, and continuity of operations is an attractive outcome of technologies and business models available today. The challenges are more related to business processes and human factors, both of which require organizational transformations to take best advantage of the collaborative environments enabled through use of cloud computing and access to broadband communications.
Gaining the most benefit from planning an interoperable environment for governments and organizations may be facilitated through use of business tools such as cloud computing. Cloud computing and underlying technologies may create an operational environment supporting many strategic objectives being considered within government and private sector organizations.
Reaching target architectures and capabilities is not a single action, and will require a clear understanding of current “as-is” baseline capabilities, target requirements, the gaps or capabilities need to reach the target, and establishing a clear transitional plan to bring the organization from a starting “as-is” baseline to the target goal.
To most effectively reach that goal requires an understanding of the various contributing components within the transformational ecosystem. In addition, planners must keep in mind the goal is not implementation of technologies, but rather consideration of technologies as needed to facilitate business and operations process visions and goals.
Interoperability and Enterprise Architecture
Information technology, particularly communications-enabled technology has enhanced business process, education, and the quality of life for millions around the world. However, traditionally ICT has created silos of information which is rarely integrated or interoperable with other data systems or sources.
As the science of enterprise architecture development and modeling, service-oriented architectures, and interoperability frameworks continue to force the issue of data integration and reuse, ICT developers are looking to reinforce open standards allowing publication of external interfaces and application programming interfaces.
Cloud computing, a rapidly maturing framework for virtualization, standardized data, application, and interface structure technologies, offers a wealth of tools to support development of both integrated and interoperable ICT resources within organizations, as well as among their trading, shared, or collaborative workflow community.
The Institute for Enterprise Architecture Development defines enterprise architecture (EA) as a “complete expression of the enterprise; a master plan which acts as a collaboration force between aspects of business planning such as goals, visions, strategies and governance principles; aspects of business operations such as business terms, organization structures, processes and data; aspects of automation such as information systems and databases; and the enabling technological infrastructure of the business such as computers, operating systems and networks”
ICT, including utilities such as cloud computing, should focus on supporting the holistic objectives of organizations implementing an EA. Non-interoperable or shared data will generally have less value than reusable data, and will greatly increase systems reliability and data integrity.
Business Continuity and Disaster Recovery (BCDR)
Recent surveys of governments around the world indicate in most cases limited or no disaster management or continuity of operations planning. The risk of losing critical national data resources due to natural or man-made disasters is high, and the ability for most governments maintain government and citizen services during a disaster is limited based on the amount of time (recovery time objective/RTO) required to restart government services, as well as the point of data restoral (recovery point objective /RPO).
In existing ICT environments, particularly those with organizational and data resource silos, RTOs and RPOs can be extended to near indefinite if both a data backup plan, as well as systems and service restoral resource capacity is not present. This is particularly acute if the processing environment includes legacy mainframe computer applications which do not have a mirrored recovery capacity available upon failure or loss of service due to disaster.
Cloud computing can provide a standards-based environment that fully supports near zero RTO/RPO requirements. With the current limitation of cloud computing being based on Intel-compatible architectures, nearly any existing application or data source can be migrated into a virtual resource pool. Once within the cloud computing Infrastructure as a Service (IaaS) environment, setting up distributed processing or backup capacity is relatively uncomplicated, assuming the environment has adequate broadband access to the end user and between processing facilities.
Cloud computing-enabled BCDR also opens opportunities for developing either PPPs, or considering the potential of outsourcing into public or commercially operated cloud computing compute, storage, and communications infrastructure. Again, the main limitation being the requirement for portability between systems.
ICT modernization will drive change within all organizations. Transformational readiness is not a matter of technology, but a combination of factors including rapidly changing business models, the need for many-to-many real-time communications, flattening of organizational structures, and the continued entry of technology and communications savvy employees into the workforce.
The potential of outsourcing utility compute, storage, application, and communications will eliminate the need for much physical infrastructure, such as redundant or obsolete data centers and server closets. Roles will change based on the expected shift from physical data centers and ICT support hardware to virtual models based on subscriptions and catalogs of reusable application and process artifacts.
A business model for accomplishing ICT modernization includes cloud computing, which relies on technologies such as server and storage resource virtualization, adding operational characteristics including on-demand resource provisioning to reduce the time needed to procure ICT resources needed to respond to emerging operational or other business opportunities.
IT management and service operations move from a workstation environment to a user interface driven by SaaS. The skills needed to drive ICT within the organization will need to change, becoming closer to the business, while reducing the need to manage complex individual workstations.
IT organizations will need to change, as organizations may elect to outsource most or all of their underlying physical data center resources to a cloud service provider, either in a public or private environment. This could eliminate the need for some positions, while driving new staffing requirements in skills related to cloud resource provisioning, management, and development.
Business unit managers may be able to take advantage of other aspects of cloud computing, including access to on-demand compute, storage, and applications development resources. This may increase their ability to quickly respond to rapidly changing market conditions and other emerging opportunities. Business unit managers, product developers, and sales teams will need to become familiar with their new ICT support tools. All positions from project managers to sales support will need to quickly acquire skills necessary to take advantage of these new tools.
The Role of Cloud Computing
Cloud computing is a business representation of a large number of underlying technologies. Including virtualization, development environment, and hosted applications, cloud computing provides a framework for developing standardized service models, deployment models, and service delivery characteristics.
The US National Institute of Standards and Technology (NIST) provides a definition of cloud computing accepted throughout the ICT industry.
“Cloud computing is a model for enabling ubiquitous, convenient, on-demand network access to a shared pool of configurable computing resources that can be rapidly provisioned and released with minimal management effort or service provider interaction.“
While organizations face decisions related to implementing challenges related to developing enterprise architectures and interoperability, cloud computing continues to rapidly develop as an environment with a rich set of compute, communication, development, standardization, and collaboration tools needed to meet organizational objectives.
Data security, including privacy, is different within a cloud computing environment, as the potential for data sharing is expanded among both internal and potentially external agencies. Security concerns are expanded when questions of infrastructure multi-tenancy, network access to hosted applications (Software as a Service / SaaS), and governance of authentication and authorization raise questions on end user trust of the cloud provider.
A move to cloud computing is often associated with data center consolidation initiatives within both governments and large organizations. Cloud delivery models, including Infrastructure as a Service (IaaS) and Platform as a Service (PaaS) support the development of virtual data centers.
While it is clear long term target architectures for most organizations will be an environment with a single data system, in the short term it may be more important to decommission high risk server closets and unmanaged servers into a centralized, well-managed data center environment offering on-demand access to compute, storage, and network resources – as well as BCDR options.
Even at the most basic level of considering IaaS and PaaS as a replacement environment to physical infrastructure, the benefits to the organization may become quickly apparent. If the organization establishes a “cloud first” policy to force consolidation of inefficient or high risk ICT resources, and that environment further aligns the organization through the use of standardized IT components, the ultimate goal of reaching interoperability or some level of data integration will become much easier, and in fact a natural evolution.
Nearly all major ICT-related hardware and software companies are re-engineering their product development to either drive cloud computing, or be cloud-aware. Microsoft has released their Office 365 suite of online and hosted environments, as has Google with both PaaS and SaaS tools such as the Google Apps Engine and Google Docs.
The benefits of organizations considering a move to hosted environments, such as MS 365, are based on access to a rich set of applications and resources available on-demand, using a subscription model – rather than licensing model, offering a high level of standardization to developers and applications.
Users comfortable with standard office automation and productivity tools will find the same features in a SaaS environment, while still being relieved of individual software license costs, application maintenance, or potential loss of resources due to equipment failure or theft. Hosted applications also allow a persistent state, collaborative real-time environment for multi-users requiring access to documents or projects. Document management and single source data available for reuse by applications and other users, reporting, and performance management becomes routine, reducing the potential and threat of data corruption.
The shortfalls, particularly for governments, is that using a large commercial cloud infrastructure and service provider such as Microsoft may require physically storing data in location outside of their home country, as well as forcing data into a multi-tenant environment which may not meet security requirements for organizations.
Cloud computing offers an additional major feature at the SaaS level that will benefit nearly all organizations transitioning to a mobile workforce. SaaS by definition is platform independent. Users access SaaS applications and underlying data via any device offering a network connection, and allowing access to an Internet-connected address through a browser. The actual intelligence in an application is at the server or virtual server, and the user device is simply a dumb terminal displaying a portal, access point, or the results of a query or application executed through a command at the user screen.
Cloud computing continues to develop as a framework and toolset for meeting business objectives. Cloud computing is well-suited to respond to rapidly changing business and organizational needs, as the characteristics of on-demand access to infrastructure resources, rapid elasticity, or the ability to provision and de-provision resources as needed to meet processing and storage demand, and organization’s ability to measure cloud computing resource use for internal and external accounting mark a major change in how an organization budgets ICT.
As cloud computing matures, each organization entering a technology refresh cycle must ask the question “are we in the technology business, or should we concentrate our efforts and budget in efforts directly supporting realizing objectives?” If the answer is the latter, then any organization should evaluate outsourcing their ICT infrastructure to an internal or commercial cloud service provider.
It should be noted that today most cloud computing IaaS service platforms will not support migration of mainframe applications, such as those written for a RISC processor. Those application require redevelopment to operate within an Intel-compatible processing environment.
Cloud computing components are currently implemented over an Internet Protocol network. Users accessing SaaS application will need to have network access to connect with applications and data. Depending on the amount of graphics information transmitted from the host to an individual user access terminal, poor bandwidth or lack of broadband could result in an unsatisfactory experience.
In addition, BCDR requires the transfer of potentially large amounts of data between primary and backup locations. Depending on the data parsing plan, whether mirroring data, partial backups, full backups, or live load balancing, data transfer between sites could be restricted if sufficient bandwidth is not available between sites.
Cloud computing is dependent on broadband as a means of connecting users to resources, and data transfer between sites. Any organization considering implementing cloud computing outside of an organization local area network will need to fully understand what shortfalls or limitations may result in the cloud implementation not meeting objectives.
The Service-Oriented Cloud Computing Infrastructure (SOCCI)
Governments and other organizations are entering a technology refresh cycle based on existing ICT hardware and software infrastructure hitting the end of life. In addition, as the world aggressively continues to break down national and technical borders, the need for organizations to reconsider the creation, use, and management of data supporting both mission critical business processes, as well as decision support systems will drive change.
Given the clear direction industry is taking to embrace cloud computing services, as well as the awareness existing siloed data structures within many organizations would better serve the organization in a service-oriented framework, it makes sense to consider an integrated approach.
A SOCCI considers both, adding reference models and frameworks which will also add enterprise architecture models such as TOGAF to ultimately provide a broad, mature framework to support business managers and IT managers in their technology and business refresh planning process.
SOCCIs promote the use of architectural building blocks, publication of external interfaces for each application or data source developed, single source data, reuse of data and standardized application building block, as well as development and use of enterprise service buses to promote further integration and interoperability of data.
A SOCCI will look at elements of cloud computing, such as virtualized and on-demand compute/storage resources, and access to broadband communications – including security, encryption, switching, routing, and access as a utility. The utility is always available to the organization for use and exploitation. Higher level cloud components including PaaS and SaaS add value, in addition to higher level entry points to develop the ICT tools needed to meet the overall enterprise architecture and service-orientation needed to meet organizational needs.
According to the Open Group a SOCCI framework provides the foundation for connecting a service-oriented infrastructure with the utility of cloud computing. As enterprise architecture and interoperability frameworks continue to gain in value and importance to organizations, this framework will provide additional leverage to make best use of available ICT tools.
The Bottom Line on ICT Modernization
The Internet Has reached nearly every point in the world, providing a global community functioning within an always available, real-time communications infrastructure. University and primary school graduates are entering the workforce with social media, SaaS, collaboration, and location transparent peer communities diffused in their tacit knowledge and experience.
This environment has greatly flattened any leverage formerly developed countries, or large monopoly companies have enjoyed during the past several technology and market cycles.
An organization based on non-interoperable or standardized data, and no BCDR protection will certainly risk losing a competitive edge in a world being created by technology and data aware challengers.
Given the urgency organizations face to address data security, continuity of operations, agility to respond to market conditions, and operational costs associated with traditional ICT infrastructure, many are looking to emerging technology frameworks such as cloud computing to provide a model for planning solutions to those challenges.
Cloud computing and enterprise architecture frameworks provide guidance and a set of tools to assist organizations in providing structure, and infrastructure needed to accomplish ICT modernization objectives.
We all know the buildings, One Wilshire, The Westin Building, 60 Hudson, Telehouse (UK), 200 Paul, 1102 Grand – all buildings advertising dozens, or even hundreds of carriers using the properties for interconnections at the fiber and network level. Meet-me-rooms are crowded, ladder racks full, and each property sits in the middle of the central business district in large cities.
At one point, rumors circled the industry that if One Wilshire had a catastrophic failure of infrastructure that global communications may be set back to the mid 1960s. True or not, the building’s meet-me-room supports hundreds of interconnections which are single-threaded connecting carriers in distant countries and continents.
All in buildings designed to house office users. All buildings with little or no potential for external security. All buildings with challenges for internal electrical distribution, cooling, and Fiber infrastructure that is not going away anytime soon.
The Internet, and what the internet is likely to evolve in the future, will be a combination of high performance wireless access and fixed access connecting every square centimeter of countries to what we are now calling the Fourth Utility. The Fourth Utility being a marriage of broadband infrastructure and cloud computing infrastructure.
As a utility, the infrastructure should be envisioned, planned, and implemented as basic infrastructure, with compromise only considered to accommodate exceptions, such as legal limitations or geological limitations. Where we need to interconnect segments of this infrastructure, as in carrier hotels, the interconnection points should be designed as infrastructure, and not a compromise.
Of course this is not surprising. Carrier hotels by design evolved from a need to find methods for competitive carriers and networks to directly interconnect without the requirement to use an incumbent or formerly monopoly carrier as a transit point. During the period of global telecom deregulation in the 1980s and 1990s those carriers scrambled to find common interconnection points near metro and long distance fiber routes.
Locations offering a neutral location in close proximity to major metro, long distance, and transcontinental submarine cable routes were found in the central business districts (CBDs) of Los Angeles, Seattle, Miami, New York, and London. Most of those locations (exception Miami/NAP of the Americas) did not have the space, nor did carriers have the money, to construct a proper central office-grade facility in the CBD to accommodate the electronic switches and muxes needed to support the carriers.
Thus the most suitable, and available, office building was selected to meet the most basic needs of the carriers.
During the 1990s global telecom deregulation progressed and changes in ownership of submarine cables allowed large numbers of international carriers to establish a presence in the carrier meet-me-rooms (MMRs) in buildings such as One Wilshire. The MMRs were operated by building landlords, with little or no telecom industry operational experience, resulting in installations which were far below normal telecom industry standards.
While MMRs have improved greatly during the past few years, the reality is we have a tremendous amount of national infrastructure being built into properties not designed for the telecom industry – infrastructure that will continue being more and more essential to our ability both as a nation, and as a member of the global economic and social community.
The United States should view telecom and cloud computing as a utility, critical to the national infrastructure. Standards that follow the same principles of roads, water, and electrical distribution must be applied to the telecom industry, including carrier hotels and other implementations contributing to the Fourth Utility.
The telecom industry must not accept new MMR or carrier hotel infrastructure that is not a design custom suited to the needs of carriers requiring interconnections. In addition, no infrastructure can tolerate single points of failure on the backbone. You cannot control a point of failure at all access points, much like an access road washing out during a flood – however the backbone must have resiliency and redundancy.
Here is a call to action for the telecom industry. Do not accept, support, contribute to, or participate in infrastructure deployments which do not provide levels of both operational and physical security needed to ensure our critical Fourth Utility of telecom infrastructure needed to protect our national and global interests.