Information Technology Reference
In-Depth Information
storage, or networks; Platform as a Service (PaaS), providing an operating sys-
tem as well as a set of tools and services to the user; or Software as a Service
(SaaS) that allows providers to grant customers with access to licensed software.
3.2 The OurGrid Middleware
OurGrid is an open source middleware for grid and cloud computing based on
a peer-to-peer architecture, developed by researchers at Universidade Federal
do Campina Grande, Brazil [5]. This middleware enables the creation of peer-
to-peer computational grids, and it is intended to speed up the execution of
Bag-of-Tasks (BoT) applications.
The OurGrid architecture is built by aggregating several participants in a grid
environment, allowing them to use remote and local resources to run their appli-
cations. OurGrid uses the eXtensible Messaging and Presence Protocol (XMPP),
an open technology for real-time communication which powers a wide range of
applications, including instant messaging, presence, multi-party chat, voice and
video calls, collaboration, lightweight middleware, content syndication, and gen-
eralized routing of XML data. XMPP allows federation, it is Internet-friendly,
and ecient, since several services can use the same XMPP server.
The main components of the OurGrid architecture are:
-
Broker
: implements the user interface to the grid. By using the broker, the
users can submit jobs to the grid and also track their execution. All the
interaction between the user and the grid infrastructure is performed through
the broker.
-
Worke r s
: used for processing the jobs submitted to the grid. Each worker
represents a real computing resource. OurGrid workers support virtualiza-
tion, and so they offer an isolated platform for executing jobs comprising no
risks to the local system running the component.
-
Peers
: have a twofold role; from the point-of-view of the user, they search
and allocate corresponding computing resources for the execution of his jobs.
From the point-of-view of the infrastructure (implicitly, for the site adminis-
trator) each peer is responsible for determining which workers can be used to
execute an application, and also how they will be used. Normally, it is enough
to have one peer per site. Communication between peers makes possible to
execute jobs remotely; in case that the local resources are not enough for
satisfying the requirements of a job, the peer seeks for additional resources
available in remote sites.
-
Discovery Service
: keeps updated information about the system composition.
It finds out the end points that peers should use to communicate with.
All these components are integrated in a transparent way to the user, allowing
the system to provide a single-image of an infrastructure with a large computing
power. A description of the OurGrid architecture is shown in Fig. 3.
