Information Technology Reference
In-Depth Information
IAAS Cloud. It defines entities, relationships API and protocols for all kinds of
management tasks. This solutions is aimed at the fulfillment of three require-
ments: integration, portability and interoperability for common tasks including
deployment, autonomic scaling and monitoring still offering an high degree of
extensibility. The OpenNebula solution already implements a RESTFull OCCI
compliant interface, and other technologies like Eucalyptus and Openstack are
working to be themselves compliant with it. Commercial providers are going to
support themselves management facilities which enable the integration of third
party clouds. For example Amazon can use its elasticity capability also exploiting
computing resources shared by OpenNebula. Rather than adopting a standard
or developing a new service interface, some efforts have been spent to develop
technologies for integration of existing IAAS Clouds. DeltaCloud and JClouds
are two different solutions. The first one provides a service with an uniform
interface, but uses different drivers to redirect requests to the supported hetero-
geneous commercial Cloud providers and to private Clouds developed by open
technologies. JCLouds, instead, offers a uniform and extensible API to develop
applications that can directly interoperate with multiple IAAS Clouds. Another
example of free and open source technology that aims at supporting the inte-
grated management of heterogeneous Cloud is provided by My Cloud Portal
1
.
It allows for setting up and managing of hybrid cloud, private and public, by
the integration of Eucalyptus and Azure Cloud infrastructures. It provides a
web interface by which it is possible to define workflows, perform monitoring
activities and reconfigure settings. Infrastructure- level resource monitoring [6]
[3] aims at the measurement and reporting of system parameters related to real
or virtual infrastructure services offered to the user (e.g. CPU, RAM, or data
storage parameters). Traditional monitoring technologies for single machines or
Clusters are restricted to locality and homogeneity of monitored objects and,
therefore, cannot be applied in the Cloud in an appropriate manner [8]. At the
state of the art there are many tools which provide Cloud monitoring facilities,
like Cloudkick, Nimsoft Monitor
2
, Monitis
3
, Opnet, RevealCloud. All of them
are proprietary solutions and do not aim at defining a standard for monitoring
interoperability. Some technologies for monitoring network and host like sFlow
have been extended in order to support the transport of monitoring information
of virtualized resources. For example host-sflow [1] exports physical and virtual
server performance metrics by using the sFlow [2] protocol. Ganglia and other
collectors of performance measures are already compliant with its specification.
In [9] authors claims that an approach based on software agents is a natural way
to tackle the monitoring tasks in the aforementioned distributed environments.
Agents move and distribute themselves to perform their monitoring tasks. In
[11] an optimal control of mobile monitoring agents in artificial-immune-system-
based (AIS-based) monitoring networks has been studied.
