10
offers some Java and Phyton based environments) to follow the entire life-cycle of appli-
cation development.
This model usually ease the cloud customer from IT administration tasks as the under-
lying infrastructure is not manageable, however cloud consumer has the control over the
application deployment and the configuration settings of the IT resources for the appli-
cation hosting.
Examples of PaaS are: AWS Elastic Beanstalk, IBM Watson IoT, Windows Azure, Her-
oku, Force.com, Google App Engine and Apache Stratos.
SaaS, or Software as a Service model, usually refers to a fully-available and pre-pack-
aged environment that cloud customers can use over cloud services. This solution allows
the customers to access to a service that is really easy and quick to setup, allowing also
the cloud provider to re-use the same cloud product for several customers. Cloud users,
in this model, do not have any administrative access and control over the IT resources,
only minimal settings changes on the software itself can be done.
Multitenancy technologies are used to distribute load on several resources, making the
SaaS a reliable and distributed service. SaaS can be both a
“pay-per-use” or a “free-of-
charge
” service for the users. In the second models the
provider would get revenues
from commercial advertisements or re-selling statistical information of the service users.
Examples of PaaS are Google Apps, Microsoft Office 365 and many other commercial
webmail platform.
Over the recent year a multitude of more specialized service models was released mostly
focusing on a specific services. Examples are:
Storage as a Service,
Database as a
Service, Security as a Service, Process as a Service, Testing as a Service, Integration
as a Service etc. Additionally also combination of cloud delivery models can be offered
to customers, for example IaaS plus PaaS can give the cloud user a software develop-
ment kit also granting a major degree of administering resources
compared to only a
PaaS scenario.
