Distributed Computing and Systems
Distributed Systems II
TDA297(CTH), DIT290(GU)
Introduction & Overview
Philippas Tsigas
About Us
Philippas Tsigas
tsigas@chalmers.se
Phone: 772 5409
Room: 5106
Philippas Tsigas
2
About Us, the TAs
– Ioannis Nikolakopoulos
ioaniko@chalmers.se
Phone: 772 5720
Room: 5107
– Ivan Walulya
ivanw@chalmers.se
Phone: 772 5720
Room: 5107
– Aras Atalar
aaras@chalmers.se
Room: 5119
– Office hours: To be
announced
3
WHY DISTRIBUTED SYSTEMS?
LET’S HAVE A LOOK AT THE EVOLUTION OF
COMPUTER SYSTEMS
Is this a Computer?
Is this a Computer?
Is this a Computer?
A modern automobile contains 50-100 ECUs and this number is increasing.
Is this a Computer?
Is this a Computer?
Is this a Computer?
Using skin to transmit information
Jul 1st 2004
From The Economist print edition
IT SOUNDS like an April Fool's Day joke, but it isn't. Microsoft, that
imperialist of the information-technology world, has actually
succeeded in patenting the human body as a computer network. US
Patent 6,754,472, issued to the company on June 22nd, is for a
method and apparatus for transmitting power and data using the
human body.…
Copyright – Facebook.com
Do You Know Where Your Data Are?
The Future is Distributed
• Bell’s Law
• Moore’s Law
log (people per computer)
– New computing class every 10
years
year
WHAT DOES INDUSTRY PROJECT?
13
Tomorrow?
Today
2020: More than 50 Billion
connected devices (Ericsson)
14
Tomorrow?
Today
2020: More than 50 Billion
connected devices (Ericsson)
15
Tomorrow?
• Distributed Systems touch all aspects of daily life!
– Integral building block for our networked society
16
WHAT DOES SOCIETY NEED?
17
Distributed Systems
• Rapidly growing interest in distributed systems
– An alternative model for the provision of socially critical
resources related fundamental services: transportation,
energy, water, etc
This is occurring in response to a deepening
awareness of the vulnerability of these services
to the challenges, such as:
• climate change (i.e., greenhouse gas emissions)
• resource scarcity (e.g., fossil fuel)
Distributed Systems
• Rapidly growing interest in distributed systems
–These
An alternative
for the
provision
of sociallyfrom
critical
are allmodel
lessons
that
were learned
resources related fundamental services: transportation,
the
unforeseen
energy,
water, etc development of the Internet!
– Simultaneously there is an increasing understanding
of how localised and networked systems can act as a
catalyst for social innovation, social connectedness,
citizen engagement and community resilience
Vehicular systems are distributed systems
Transportation can dramatically
improve via distributed coordination
Vehicular systems are expected to gear vehicles
with autopilot capabilities, improve safety, reduce
energy consumption, lessen CO2 omission…
Currently, these solutions are not allowed
to operate on public roads, as the risk of
causing severe damage cannot be
excluded with sufficient certainty
Inherently Uncertain Environment
Environment
Communications
User
Highly Available
Complex Control
How can we affordably validate the
safety properties of future largescale vehicular systems?
Control
Sensors
The SmartGrid is a Distributed System
- The electricity grid facilitated the innovations of
the 20th century
- The SmartGrid is to support the green advances
of the 21st. Without it, much of the other green
technology cannot work
The SmartGrid is a Distributed System
improve and protect these systems through:
• adaptive coordination for distribution/allocation => energyefficiency: e.g. microgrids, virtual powerplants
24
THE COURSE STRUCTURE
25
Reading
• 5th Edition of the book: "Distributed Systems:
Concepts and Design"
– written by: George Coulouris, Jean Dollimore and
Tim Kindberg, Gordon Blair
– published by Addison-Wesley, ISBN 0-13-214301-1
• Some extra material: Distributed Algorithms
(Notes) + papers.
Description
Distributed systems are popular and powerful computing
paradigms. Their importance increases as networked
computers become more common than freestanding
ones, especially since many different types of computers
can be found in networks. In this course we will see the
points of inherent difference and strength of distributed
systems compared with sequential or strongly-coupled
systems; consequently, we will also study the issues and
problems that have to be addressed and solved efficiently
for these differences to be taken advantage of, so that the
system retains its strength and high potential.
Course Goals
1.Replication; The advantages and costs of replicating
data:
– Potential improvement in response times and reliability
– Extra communication costs involved in keeping data
consistent.
2.Fault-tolerant Agreement in Distributed Systems:
– (a very special and significant problem, since it is a key
issue in most synchronisation and coordination problems
in distributed systems)
– Study of the basic impossibility results and discuss their
implications
– Proceed with solutions and protocols for systems with
certain strengths and design structures.
Goals cont.
3. Resource Allocation:
Scalable and Fault-tolerant (Local) Resource
Allocation
4. Distributed algorithms:
How to design and analyse distributed
algorithms?
5. Sensor Networks:
Energy issues
Course Goals in a Nutshell
• Lectures: Teach you to analyze and understand
how Basic Components of Distributed Systems
work.
• Labs: Give you hands-on experience
– Feel the challenges
– Master the techniques
• Have some fun!
– Optional: you can pass this course without
30
Approximate Course Schedule
32
Resources
Course Homepage
• http://www.cse.chalmers.se/edu/course/TDA297/
• http://pingpong.chalmers.se/public/courseId/3499/langsv/publicPage.do?item=1690031
Slides:
• At homepage after lecture
• Last years slides available, use them as a reference point before the
lecture:
• http://pingpong.chalmers.se/public/courseId/2284/langsv/publicPage.do?item=1119047
• http://www.cse.chalmers.se/edu/year/2013/course/TDA297/
Full support page for the Coulouris’ book:
• http://www.cdk5.net/
Labs
1st Assignment:
• Distributed bulletin board
URL from the previous year:
http://pingpong.chalmers.se/public/courseId/2284/lan
g-sv/publicPage.do?item=1118129
New page coming up soon
Labs cont(2).
2nd Assignment:
• Reliable and ordered multicast
http://pingpong.chalmers.se/public/courseId/2284/lan
g-sv/publicPage.do?item=1118978
3nd Assignment:
• Routing in Sensor Networks
http://pingpong.chalmers.se/public/courseId/2284/lan
g-sv/publicPage.do?item=1118978
Homework assignment (Optional)
• You will get up to 5 extra points for the exam
depending on the quality of your work
• Select a paper related to one of the specified
topics, write a report of 2-3 pages and present
it in 15-20 minutes (including 5 minutes of
discussion)
• The report and presentations include the
explanation of main ideas in the paper
together with some background information
36
To Pass
• Pass all 3 labs tasks
– Not graded
– 2 students per group, if possible
• Pass written exam
– Graded
• (extra points from the Homework assignment)
38
Lectures: Please Come to Class
• Lectures deepen and discuss
– Difficult to do from just slides or book
• This is key!
– Ask questions
• Do not be shy
• In general
– Statistically shown: People who come to class have
• Better grades
• A higher probability of passing
39
Labs: Deadlines
• Submission: submit your final solution
– Each submission will be graded: pass / no pass
– Feedback: At the “Assignment Lectures: …
+ Questions and answers for all labs.”
40
Questions?
41