Cloud Computing and Agile Organization Development

In the 3 millennium economy, defined by globalization and continuous reduction of natural resources, the economic organization becomes the main actor in the phenomenon of transformation and adaptation to new conditions. Even more, the economic environment, which is closely related to the social environment, undergoes complex metamorphoses, especially in the management area. In this dynamic and complex social and environmental context, the economic organization must possess the ability to adapt, becoming a flexible and agile answer to new market opportunities. Considering the spectacular evolution of information and communications technology, one of the solutions to ensure organization agility is cloud computing. Just like the development of any science requires adaptation to theories and instruments specific to other fields, a cloud computing paradigm for the agile organization must appeal to models from management, cybernetics, mathematics, structuralism and information theory (or information systems theory).


Objectives
This analytical-methodological and theoretical-practical endeavour constitutes an original scientific incursion into the philosophy and theory of cloud computing from the perspective of turning it into a real framework for ensuring agility in the development of economic organizations.Also, theoretical approach is supported by highlighting practical aspects regarding cloud computing market and the influence big actors of this market may have on the success of a cloud computing paradigm for organization development.The original approach brings forth the emergent production factors of the knowledge society, digital economy and globalization phenomenon.They include information, organization culture, and enterprise architecture in terms of business, information and technology.

Cloud computing philosophy
In the spring of 2001, in Claremont, California, Peter F. Druckerconsidered one of the founders of modern science of management by business specialists and academicstalking about the "challenge ahead", said "we may almost certainly say that the challenges we will have to face in the tomorrow's conditions are of management nature and concern the individuals.… those tasks cannot be performed by the state, but only by individual organizationslucrative enterprises and nongovernmental non-lucrative organizationsand individuals" [1] Tomorrow's economic conditions invoked by Drucker are visible even today and the socialeconomic entities mentioned are indeed the economic organizations and the individuals.The management philosophy of the contemporary organization includes both global market agility and modern paradigms that target directly the human resource (as primary production factor), like the anthropocentric model.Even more, when the agility approach is supported by the use of information and communications technology instruments, cloud computing becomes its structural foundation.Cloud computing is a relatively modern ICT concept and represents an aggregation of distributed services for computation, applications, data access and storage.The cloud computing philosophy assumes that the entire ensemble is hidden to the user, which does not know the location and physical configuration of the systems that provide the services.Terminologically, cloud computing derives from a symbolic graphical representation of the 1 DOI: 10.12948/issn14531305/18.4.2014.01global network as a cloud, used when technical details of the internet may be ignored (Figure 1).In other words, cloud computing represents one of the significant tendencies of ICT evolution, bringing a new model for provision, management and security within the organization.Currently, there is no universally accepted definition of cloud computing.The national Institute of Standards and Technology (NIST) and Cloud Security Alliance (CSA) present a base for the definition, describing cloud computing as a model that provides access on demand, through a network to an array of shared configurable information resources (like networks, servers, storage space, applications and services).These are available immediately and with little management effort or involvement from the service provider.The main attributes that support cloud computing concept and philosophy from the perspective of using it as an infrastructure for agile organization development [3] are:  abstract computation and information technology service oriented approach;  virtual, dynamic, scalable and massive infrastructure;  shared, configurable, flexible and dynamic resources;  access via internet to all devices;  platform with little or no management at all (self-managed);  employment model based on self-service;  billing based on consumption (measured/metered service).The latest professional studies available highlight that over 70% of the economic organizations consider cloud computing a real technological option, while about 70% consider adopting it will lead to the flexibility of the business.Even more, over 60% of the organizations consider the cloud computing paradigm will lead to increased reaction speed of the business to the new market conditions, while 65% estimate that it will lead to increased focus on the main aspects of the business.[4] Although there is a real growth trend, we may say that currently very few organizations use cloud computing, mainly due to lack of correct understanding of the architecture, payment model, different business environment and specific elements of each organization.[5] 3 Cloud computing technology Built on the attributes supporting its philosophy, cloud computing technology consists of the convergence of five main vectors [6]: on demand service (the client may unilaterally reserve cloud resources); remote network access (heterogeneous clients access cloud resources DOI: 10.12948/issn14531305/18.4.2014.01 through global internet); shared resources (cloud provider dynamically (re)allocates resources to clients according to their demands); capacity to scale resources (rapid scaling in both directions of cloud resources available to clients); service metering (implementation of a pay-per-use system).From the technological perspective of the provided service, NIST defines three main models of specific cloud computing services (this approach is unanimously accepted): software as service (SaaS), platform as service (PaaS) and infrastructure as service (IaaS) -Figure 2.
 Saas -Software as a serviceis a model that allows the use of applications through a cloud infrastructure; the applications are available through heterogeneous devices, using a web-based interface (for example webmail); users cannot control or manage the infrastructure offered by the cloud provider.Being one of the most used cloud services, the list of applications that may work in SaaS regime includes enterprise services (ERP), digital signature, CRM applications, management applications (specific to agile organizationsfinancial, human resources, sales, project management), search instruments etc.This model is recommended when the data is confidential to the organization.
 PaaS -Platform as a Serviceis a model that allows users to develop or use applications using programming languages provided by the cloud or created by the user; access to existing or created applications as well as platform configuration elements/services (session management, device integration, testing environments, content management etc.) can be controlled.In this model the applicability is limited considering application portability or the need to adjust the hardware-software combination to improve application performance.The model is useful for developing collaborative solutions by using cloud shared resources.
 IaaS -Infrastructure as a Serviceis a model that provides clients the possibility to reserve processing, storage, network resources (which may be used to run any software, including operating systems).Cloud clients/consumers do not control or manage the cloud infrastructure but control the operating system, storage space, installed applications and some network components (like computer-level firewalls).Services available in this model include: server hosting, web server, operating systems, bandwidth reservation etc.Generally, IaaS is employed when the organization does not want to invest in infrastructure and the growing business needs require significant ICT resources.Beyond the NIST standard, with certain limitations and on a smaller scale, considering the rendered service, there are four sub-types of cloud computing:  STaaS -STorage as a Service;  DTaaS -DeskTop as a Service;  DPaaS -Data Protection as a Service;  BPaaS -Business Process as a Service.Considering the type of infrastructure hosting, there are four ways cloud computing is implemented: public cloud (infrastructure is provided to the clients, but it is owned by the cloud service provider); community cloud (shared infrastructure between several organizations and supports a community with common interests; managements is carried out by the community or an external entity); private cloud (the infrastructure is exclusively used by a single organization, management being provided internally or by an external organization); hybrid cloud (the infrastructure is composed by two or more public, community or private clouds interconnected to ensure the portability of data and applications).DOI: 10.12948/issn14531305/18.4.2014.01

Fig. 2. Types of services provided by cloud computing
Each organization chooses from the diversity of cloud solutions available today the one that suits best its interests.From the point of view of economic organization agility, main choices are private or hybrid cloud (in spite of significantly higher costs).The factors that significantly influence the decision to adopt cloud on a large scale, and implicitly the ripening of this concept, which is bound to revolutionize business environment in the near future, are security concerns, ensuring conformity with current accepted interoperability standards, integration of cloud service with internal organization systems.

Cloud computing market
Initially, cloud computing solutions were designed for small and medium organizations, providing them with a way to increase agility with low costs.Lately, more and more large organizations analyze the possibility of quick integration of new solutions and use of cloud computing.Still, cloud computing remains best suited for small and medium organizations (Figure 3).One of the key findings is that over 60% of the organizations use cloud services and the growth rate is about 10%.

Organization agility: limits of cloud computing model
In a continuously changing global market, organization agility is a key element in gaining strategic advantage.This may be ensured by achieving an agile architecture of the organization, lowering the time for development of new processed and increasing flexibility of existing processes.The need for agility and ensuring the versatile character of the organization are premises for lowering the time to answer client requests and extend the client pool.Thus, the organization may account lower costs for adaptation to market changes and therefore increased revenue.Agile organization requires agile architectures, techniques, methods and instruments in order to react fast to change requirements.The path towards an agile business requires continuous control of own business process dynamics, human resources and informational system.The literature and economic practice records three main approaches (still in use today) for achieving organization agility:  Business Process Management (BPM) to orchestrate independent functionalities;  Service Oriented Architecture (SOA) for DOI: 10.12948/issn14531305/18.4.2014.01 the architecture of design and implementation of these functionalities;  Decision Management (DM) for management of organization decisions.In the context of cloud computing paradigmas an original element of scientific researchthe enterprise architecture must be perceived as a rigorous description of its structure, decomposition in subsystems, relations between systems, relations with the external environment, and principles that must be observed during design and evolution of the organization.There are three architectural levels for a level 21st economic organization: business architecture, information architecture and technological architecture.Considering the meaning of each architectural level, we can fiind significant elements of the cloud computing paradigm at least in the informational and technological levels (Figure 4).

Fig. 4. Architecture levels in cloud computing
Thus, the business architecture will exhaustively and systematically describe the main business processes of the organization and the information architecture will describe the applications, data and the way they integrate.Technological architecture concerns technologies accepted for operating platforms, networks, various applications, collaboration, data representation and manipulation, integration, security and systems management.Beyond the positive aspects of the cloud computing paradigm for agile organization development, we must consider the limitations of this model.They are mainly related to confidentiality issues.Once the data and applications are placed in a cloud, there is a risk that the parties that have access (cloud computing providers for examplesoftware, hosting, and communications) will easily monitor every operation (unfortunately there are persons with malicious intents).Even more, there are legal risks: for example, not knowing where the business data is physically stored may lead in involuntary export/import of data or software.What is legal in a country (national legislation) may be illegal where the data is stored.Also, the organization that adopts cloud computing may be exposed to security risksthere is a significant number of risks assumed when the business data are "somewhere", on a hard disk in an unknown country, where persons beyond control may access them.There are also availability problems: to a certain extent, redundancy may prevent it and ensure access to data at any moment, but there may be cases when the cloud computing provider can be forced to fully stop the operations (bankruptcy, official inquire etc.) completely cutting the organization from accessing its own data and applications.Maybe future solutions will erase these limitations and concerns.DOI: 10.12948/issn14531305/18.4.2014.01 6 Organization agility through cloud computing Implementation/adoption of the proposed cloud computing paradigm to achieve economic organization agility must be carried out within the boundaries of the architecture principles (table 1), which concern mainly the quality standards (for management, hardware, software etc.).Also, an important role belongs to the specific frameworksdescriptions of the architecture components as well as external components belonging to the environment where the architecture is designed, developed and implemented (clients, market, industry, opportunities, competition, authorities).
Cloud service models (as form of implementation) must be tuned to the specifics of architecture models, in terms of principles, services, standards, concepts, components, visualising models and configurations).Among the architectural models the following stand out: Zachman model as taxonomy for business information systems (patented model); TO-GAF (The Open Group Architecture Framework) as model for organization/owner; FEAF (Federal Enterprise Architecture Framework) and DoDTRM (Department of Defense Technical Reference Model) for defense industry [7].

Table 1 .
Architectural principles for clouds  Lean Software Development (LSD)