Sie sind auf Seite 1von 11

Distributed System

Chonnam National University


School of Electronics and
Computer Engineering

Kyungbaek Kim

Syllabus
Instructor

Kyungbaek Kim
Office : engineering building #6-715
Email : kyungbaekkim@chonnam.ac.kr
Office Hours : Wed 12:00 ~ 12:30

Lectures : Wed 09:00 ~ 12:00


Location : engineering building #7-327
Course Web Page : e-Class page on JNU Portal
Other page :
http://myweb.jnu.ac.kr/~kbkim/jnu_classes/2016f_to
pics_in_distributed_systems_index.html

Syllabus (contd)
Reference Book
Hadoop: The Definitive Guide, by Tom White
Do it! , by

Distributed systems principles and


paradigms 2nd edition, by andrew S.
Tanenbaum and Marrten Van Steen

Grading Policy
Attendance 10%
100% attendance will get bonus points

Reading Assignments and Homework 30%


Tentatively scheduled 13 papers for reading : 10%
Hadoop exercises : 20 %
Late submission takes disadvantages

Project 40%
Personal Project : 25%
Research on Hadoop related projects
Present during lectures
Two times per student

Paper presentation : 15%


One paper per student

Final Exam 20%


Team Project.
Final Presentation with a final report
Tentatively Scheduled on the last week of this semester

Reading Assignments
Each student should provide summary
reports for two papers per week
Tentatively scheduled for 13 papers

Dont do Copy and Paste of paper


contents
Try to write a summary in your words
Papers and a summary template will be
provided through e-Class

List of Reading Assignments

9/21 [2004 OSDI]MapReduce Simplified Data Processing on Large Clusters


9/28 [2010 ICDE]Hive a petabyte scale data warehouse using hadoop
10/12 [2010 HPDC]Twister A Runtime for Iterative MapReduce
10/19 [2012 HotCloud]Discretized Streams An Efficient and Fault Tolerant Model For
Stream Processing on Large Clusters
10/26 [2011 MIDDLEWARE]Resource Provisioning Framework for MapReduce Jobs with
Performance Goals
11/02 [2013 MIDDLEWARE]FlowFlex Malleable Scheduling for Flows of MapReduce Jobs
11/09 [2011 VLDB Endowment]CoHadoop Flexible Data Placement and Its Exploitation
in Hadoop
11/16 [2012 VLDB Endowment]M3R Increased Performance for In Memory Hadoop Jobs
11/23 [2012 VLDB]Muppet MapReduce Style Processing of Fast Data
11/30 [2013 BigData]Scalable Distributed Event Detection for Twitter
12/7 [2013 ICDCS]Efficient Geo-Distributed Data Processing with Rout
12/14 [2013 ICDCS]HybridMR A Hierarchical MapReduce Scheduler for Hybrid Data
Centers

Personal Project
Research on a Hadoop related project
http://projects.apache.org/
Big-Data and Database

One student takes two topic from two sets of topics


What should be provided by each student
Report : 1 page of description of the topic(IEEE
conference format without abstract)
Motivation, Definition, How to work, Expectation.

Presentation Slide
Manual of a simple Exercise to understand the project
with valid codes
Presentation (during classes)

List of topics for Personal Project

Apache
Apache
Apache
Apache
Apache
Apache
Apache
Apache

HBase
Hive
Cassandra
Zookeeper
Sqoop
Pig
Ambari
Avro

Apache
Apache
Apache
Apache
Apache
Apache
Apache
Apache

Tez
Storm
Flume
Kafka
Spark
Zeppelin
Giraph
Mahout

Team Projects
Two students make one team
Build a Map/Reduce computation module
for analyzing data which are generated
from a computer system
You need to imagine a computer system
You need to define the type of data
You need to set the target application which
will use the output of Map/Reduce computation

Required Items
Final presentation
Final report
Source Code

After this class, you will


Understand important aspects of
distributed systems and Map/Reduce
system
Explore the state of the art research
mostly related to
Map/Reduce architecture
Performance of Map/Reduce architecture
Combination between Map/Reduce and
Machine Learning

Lecture Schedule

Introduction of Distributed System


Introduction of Map/Reduce
Bigdata and MapReduce
Hadoop MapReduce
Exercises

Hadoop related Projects


Project final presentation

Das könnte Ihnen auch gefallen