technology – Cloud Computing Best Practices

By Jennifer Marsh

Jennifer Marsh is a software developer, programmer and technology writer and occasionally blogs for Rackspace Hosting.

Businesses basically have two choices for operating systems when shopping around for cloud servers: Windows or Linux. While Linux is cheaper and runs on many enterprise servers, businesses that run internal applications for a Windows desktop can benefit from Windows cloud servers.

The IT department and users will understand the platform more easily than were they to learn Linux. But launching an intuitive platform is only one of the advantages of cloud servers in a Windows environment.

Multiplatform Support

Business that have been online for several years probably have some legacy code in use in various departments. Fortunately, cloud servers can support multiple platforms for businesses moving towards a Windows platform.

Integration with Microsoft Azure

The latest Windows Server 2012 includes a cloud feature called Azure. Azure gives businesses the tools to create platforms as a service (PaaS) and integrates cloud server technology within an internal network. To take advantage of the Azure service, the business must setup a cloud hosting environment. Azure is more easily integrated with a corresponding Windows cloud host. The IT manager can use Microsoft’s wizard to install and configure the Azure server for cloud hosting.

More Cost Efficient for Support

Because most IT infrastructures have a lot of moving parts, system errors, downtime and desktop support can be expensive, especially when hosted internally. Having onsite personnel for any company can be expensive, and too little support can cost the company money. Hosting Windows services in the cloud eliminates much of the cost of having onsite support staff available seven days a week, 24×7. Check the contract for specifics before signing up for any particular service.

Additionally, hosting in the cloud means the company only pays for the bandwidth and server resources used each month and not a flat fee amount. Any cloud host charging a flat fee is in fact not a true cloud host. By paying for only what is used, businesses can cut down on IT infrastructure costs. As the business grows and more revenue is brought in, the cloud costs will also grow, but these costs only grow with the business’ success.

Open Data is data that can be freely used, reused and redistributed by anyone – subject only, at the most, to the requirement for attributes and sharealikes (Open Software Service Definition – OSSD). As a consequence, Open Data should create value and might have a positive impact in many different areas such as government (tax money expenditure), health (medical research, hospital acceptance by pathology), quality of life (air breathed in our city, pollution) or might influence public decisions like investments, public economy and expenditure. We are talking about services, so open data are services needed to connect the community with the public bodies. However, the required open data should be part of a design and then integrated, mapped, updated and published in a form, which is easy to use. MaaS is the Open Data driver and enables Open Data portability into the Cloud.

Introduction
Data models used as a service mainly provide the following topics:

Implementing and sharing data structure models;
Verifying data model properties according to private and public cloud requirements;
Designing and testing new query types. Specific query classes need to support heterogeneous data;
Designing of the data storage model. The model should enable query processing directly against databases to ensure privacy and secure changes from data updates and review;
Modeling data to predict usage “early”;
Portability, a central property when data is shared among fields of application;
Sharing, redistribution and participation of data among datasets and applications.

As a consequence, the data should be available as a whole and at a reasonable fee, preferably by finding, navigating and downloading over the Cloud. It should also be available in a usable and changeable form. This means modeling Open Data and then using the models to map location and usage, configuration, integration and changes along the Open Data lifecycle.

What is MaaS
Data models can be shared, off-line tested and verified to define data designing requirements, data topology, performance, placement and deployment. This means models themselves can be supplied as a service to allow providers to verify how and where data has to be designed to meet the Cloud service’s requisites: this is MaaS. As a consequence by using MaaS, Open Data designers can verify “on-premise” how and why datasets meet Open Data requirements. With this approach, Open Data models can be tuned on real usage and then mapped “on-premise” to the public body’s service. Further, MaaS inherits all the defined service’s properties and so the data model can be reused, shared and classified for new Open Data design and publication.

Open Data implementation is MaaS (Model as a Service) driven
Open Data is completely supported by data modeling and then MaaS completely supports Open Data. MaaS should be the first practice, helping to tune analysis and Open Data design. Furthermore, data models govern design, deployment, storage, changes, resources allocation, hence MaaS supports:

Applying Best Practice for Open Data design;
Classifying Open Data field of application;
Designing Open Data taxonomy and integration;
Guiding Open Data implementation;
Documenting data maturity and evolution by applying DaaS lifecycle.

Accordingly, Maas provides “on-premise” properties supporting Open Data design and publication:
1)    Analysis – What data are you planning to make open? When working with MaaS, a data model is used to perform data analysis. This means the Open Data designer might return to this step to correct, update and improve the incoming analysis: he always works on an “on-premise” data model. Analysis performed by model helps in identifying data integration and interoperability. The latter assists in choosing what data has to be published and in defining open datasets;
2)    Design – During the analysis step, the design is carried out too. The design can be changed and traced along the Open Data lifecycle. Remember that with MaaS the model is a service, and the data opened offers the designed service;
3)    Data security – Data security becomes the key property to rule data access and navigation. MaaS plays a crucial role in data security: in fact, the models contain all the infrastructure properties and include information to classify accesses, classes of users, perimeters and risk mitigation assets. Models are the central way to enable data protection within the Open Data device;
4)    Participation – Because the goal is “everyone must be able to use Open Data”, participation is comprehensive of people and groups without any discrimination or restriction. Models contain data access rules and accreditations (open licensing).
5)    Mapping – The MaaS mapping property is important because many people can obtain the data after long navigation and several “bridges” connecting different fields of applications. Looking at this aspect, MaaS helps the Open Data designer to define the best initial “route” between transformation and aggregation linking different areas. Then continually engaging citizens, developers, sector’s expert, managers … helps in modifying the model to better update and scale Open Data contents: the easier it is for outsiders to discover data, the faster new and useful Open Data services will be built.
6)    Ontology – Defining metadata vocabulary for describing ontologies. Starting from standard naming definition, data models provide grouping and reorganizing vocabulary for further metadata re-use, integration, maintenance, mapping and versioning;
7)    Portability – Models contain all the properties belonging to data in order that MaaS can enable Open Data service’s portability to the Cloud. The model is portable by definition and it can be generated to different database and infrastructures;
8)    Availability – The DaaS lifecycle assures structure validation in terms of MaaS accessibility;
9)    Reuse and distribution – Open Data can include merging with additional datasets belonging to other fields of application (for example, medical research vs. air pollution). Open Data built by MaaS has this advantage. Merging open datasets means merging models by comparing and synchronizing, old and new versions, if needed;
10) Change Management and History – Data models are organized in libraries to preserve Open Data changes and history. Changes are traced and maintained to restore, if necessary, model and/or datasets;
11) Redesign – Redesigning Open Data, means redesigning the model it belongs to: the model drives the history of the changes;
12) Fast BI – Publishing Open Data is an action strictly related to the BI process. Redesigning and publishing Open Data are two automated steps starting from the design of the data model and from its successive updates.

Conclusion
MaaS is the emerging solution for Open Data implementation. Open Data is public and private accessible data, designed to connect the social community with the public bodies. This data should be made available without restriction although it is placed under security and open licensing. In addition, Open Data is always up-to-date and transformation and aggregation have to be simple and time saving for inesperienced users. To achieve these goals, the Open Data service has to be model driven designed and providing data integration, interoperability, mapping, portability, availability, security, distribution, all properties assured by applying MaaS.

References
[1] N. Piscopo – ERwin^® in the Cloud: How Data Modeling Supports Database as a Service (DaaS) Implementations
[2] N. Piscopo – CA ERwin^® Data Modeler’s Role in the Relational Cloud
[3] N. Piscopo – DaaS Contract templates: main constraints and examples, in press
[4] D. Burbank, S. Hoberman – Data Modeling Made Simple with CA ERwin^® Data Modeler r8
[7] N. Piscopo – Best Practices for Moving to the Cloud using Data Models in theDaaS Life Cycle
[8] N. Piscopo – Using CA ERwin^® Data Modeler and Microsoft SQL Azure to Move Data to the Cloud within the DaaS Life Cycle
[9] The Open Software Service Definition (OSSD) at opendefinition.org

Disclaimer – This document is provided AS-IS for your informational purposes only. In no event the contains of “MaaS (Model as a Service) is the emerging solution to design, map, integrate and publish Open Data” will be liable to any party for direct, indirect, special, incidental, economical (including lost business profits, business interruption, loss or damage of data, and the like) or consequential damages, without limitations, arising out of the use or inability to use this documentation, regardless of the form of action, whether in contract, tort (including negligence), breach of warranty, or otherwise, even if an advise of the possibility of such damages there exists. Specifically, it is disclaimed any warranties, including, but not limited to, the express or implied warranties of merchantability, fitness for a particular purpose and non-infringement, regarding this document use or performance. All trademarks, trade names, service marks and logos referenced herein belong to their respective companies/offices.