Team Members Lora zalmover Roni Brodsky Academic Advisor
Professional Advisors Dr. Natalya Vanetik Prof. Shlomi Dolev Dr.
Guy Tel-Zur
Slide 2
Introduction Solution Motivation System Requirements System
Architecture User Interface
Slide 3
Cloud computing is where data, software and applications, or
computer processing power are accessed from a cloud of online
resources. Cloud computing today There are commercial tools that
provide Cloud computing services, like EC2 of Amazons. Also
existing an open-source tools like Eucalyptus which enables users
run programs on their own resources and provides a detailed view of
what would otherwise be the black box of cloud-computing
services.
Slide 4
Motivation for cloud computing enables individual users to
access their data or applications from any device. allows the
organizations to reduce their capital costs by purchasing hardware
and software as a utility service. Rather than purchasing servers,
software, data center space or network equipment, clients instead
buy those resources as a fully outsourced service. Fault tolerance
Fault-tolerance is the property that enables a system (often
computer-based) to continue operating properly in the event of the
failure of (or one or more faults within) some of its
components.
Slide 5
In our project we established a cluster of cloud computing. We
used an open-source program for managing cloud computing called
"Eucalyptus ". On top of the cloud we developed an algorithm for
fault tolerance, which purpose is to identify errors during the
execution of a program, and handle them with Fault-tolerance by
replication. Replication means providing multiple identical
instances of the same system or subsystem, directing tasks or
requests to all of them in parallel, and choosing the correct
result on the basis of a quorum. After identifying the failure our
system tries to recover several times.
Slide 6
In order to demonstrate the motivation of our system, we will
present an example. Suppose you have a statistical algorithm that
determines whether a given number is prime. Since the result is
statistical, there is some probability for getting wrong answer.
Our system significantly increases the chances of getting the right
answer. Several replicas will run the same algorithm
simultaneously, and the answer will be determined by the
majority.
Slide 7
Main functional requirements: Running a program without support
of fault tolerance on cloud computing cluster Running a program
with support of fault tolerance on cloud computing cluster
(detection and recovery) Statistical information
Slide 8
Slide 9
Server 1Server 2Server 3... Server i Cloud Client Server FTDR
Application Client