Skip to main content

Multi-Tenancy e Cloud Computing


Due to the traditional software offering "On Premisse" where the application is installed and configured to operate in "data centers" of companies, subject to specific contracts use license, we have a wide range of offerings frameworks, platforms, architecture models, "patterns" and professionals available. The SAAS model has required new architectures specifically designed for Cloud Computing, along with these drawings certain terms gain new status to identify models and technology needs of the moment. One of these terms is the "Multi-Tenancy". The model "Multi-Tenancy" is a model architecture that enables resource optimization of infrastructure and software systems through sharing retaining tenants, businesses / customers, logically separated. At the same time it is desired in supplier´s side a pattern of application / service for highly scalable applications. "Multi-Tenancy" applications typically need to be able to provide a high degree of customization to support the needs of each organization cliente that is consuming the service. Customization typically includes aspects of extensions of its objects, attributes and operations, processes, access models and even themes / styles for the enterprise environment. According Cezar Taurion Manager New Technologies / Technical Evangelist at IBM Brazil we can classify in this models:


  1. Multi-Tenancy as isolated Tenant: In this model there is no sharing of resources. In practice, although the user feel the experience of multi-tenant, because the application is offered to multiple customers from the same data center, this model is not multi-tenant because typically each tenant requires its own instance of hardware and software. Example: Keeeping each user with their own set of computational resources and its own instance of the application. 
  2. Multi-tenant via shared hardware (virtualization): similar to the previous model, in each tenant has its own set of technology dynamically allocated resources through virtualization. Model suitable for reuse of applications not designed for this. This model has limited elasticity for each tenant has his universe in a virtual machine, minimum unit of allocation and release of funds. 
  3. Multi-tenant via container: In this model, multiple tenants are running on the same instance of an application container (an application server), but each tenant is associated with a separate instance of the database. The isolation of the database ensures data integrity of tenants, while the container implementation is  shared by tenants, this model offers the advantages of elasticity. The customization is left to the software architecture. 
  4. Multi-tenant via the entire software stack shared: in this model, as well as the application container, also a single instance of the database is shared by all tenants. This model generate a concern because it needs extra security against accidental or intentional access to data from other companies. 


Thinking of economy of scale model choice "Multi-Tenancy" is essential for SAAS providers, where the cost per tenant should see minimized and agility and elasticity are essential attributes of a Cloud Computing offering. Observing the classifications above of Cezar Taurion the  models Multi-tenant via container and Multi-tenant via any software stack shared arise as models most suitable for this need.

Comments

  1. Hello Michale, messages like yours are inspiration to continue, thanks

    ReplyDelete

Post a Comment

Thanks for your attention and feedback.
Nivio dos Santos

Popular posts from this blog

CÓDIGOS UNICODE PARA CARACTERES ESPECIAIS

A/a á \u00e1 à \u00e0 â \u00e2 ã \u00e3 ä \u00e4 Á \u00c1 À \u00c0 Â \u00c2 Ã \u00c3 Ä \u00c4 E/e é \u00e9 è \u00e8 ê \u00ea ê \u00ea É \u00c9 È \u00c8 Ê \u00ca Ë \u00cb I/i í \u00ed ì \u00ec î \u00ee ï \u00ef Í \u00cd Ì \u00cc Î \u00ce Ï \u00cf O/o ó \u00f3 ò \u00f2 ô \u00f4 õ \u00f5 ö \u00f6 Ó \u00d3 Ò \u00d2 Ô \u00d4 Õ \u00d5 Ö \u00d6 U/u ú \u00fa ù \u00f9 û \u00fb ü \u00fc Ú \u00da Ù \u00d9 Û \u00db Consoantes ç \u00e7 Ç \u00c7 ñ \u00f1 Ñ \u00d1 Símbolos & \u0026 ' \u0027 Ʃ \u01a9 ° \u00b0 ª \u00aa ° \u00b0 ➕ \u2795 ➖ \u2796 ➗ \u2797 ✓ \u2713 ✗ \u2717

Parsing XML with namespace in Java

Introduction Yesterday, again, I needed parser a XML stream using Java. Then I went to Google try to find a exemple code. Because I needed the value of two specific fields, TAGs, I decided use XPATH to get directly this values. How ever, I needed do several tried until get a result. Keywords JAVA, XML, NAMESPACE, XML PARSE; Article I will show how to read XML without namespace and after we will read a XML streams with namespaces. This is the first XML that I have used in this example without namespaces <?xml version="1.0" encoding="UTF-8"?> <root>   <head>     <title>My Book</title>   </head>   <bookreview>     <title>This is a book about special girls</title>     <girls>       <girl>         <name>Mata Hari</name>       </girl>       <girl>          <name>Belle de Jour</name>       </girl>       <girl>          <name&

Neurônio Artificial Conceito

Outro dia conversando sobre redes neurais, consciência, inteligência artificial, percebi que estamos tão cercados e acostumados com tecnologia que em muitos cenários já estamos vivenciando a “Era da magia” onde as coisas funcionam mas o conhecimento de como funcionam está se perdendo. As máquinas acabam nos passando a sensação de serem seres vivos, simulando inteligência e comportamentos. Esta postagem descreve o conceito de um neurônio artificial para dar uma base às próximas postagens, porém com uma abordagem mais palpável, sem usar chips, transistores, ou outros elementos tecnológicos que fogem do conhecimento comum. O modelo aqui apresentado é um exercício mental, conceitual, fazendo uma série de simplificações e removendo elementos que só tornariam a visualização mais complexa. Para entender o exemplo descrevo aqui dois elementos: Relé, é um interruptor eletromecânico. A alavanca de contato se movimenta e fecha o circuito quando uma corrente elétrica percorre as espi