1. Distributed Systems 1. Einführung
Simon Razniewski
Faculty of Computer Science
Free University of Bozen-Bolzano
A.Y. 2016/2017

2. Lecturer Lab Assistant
Simon Razniewski
Florian Hofer
Faculty of Computer Science
FUB – POS 1.04
Piazza Domenicani 3
Office hours:
Thursday 13:00-14:00
KRDB Research Center Faculty of Computer Science FUB – POS 2.08 Piazza Domenicani 3 Office hours: Thursday 8:30-10:30

4. Me Diplom Informatik from TU Dresden, 2010 PhD from FUB, 2014
Since then researcher in the KRDB group
Involved in research on distributed querying and knowledge fusion
Taught this course 2015 and 2016
Enjoyed a lot, because DS are ubiquitous, interesting and easy

5. You
Survey!

6. Why to study DS? Learn how to make resources accessible
printers, storage, compute power, web pages

7. Why to study DS? (2)
Learn to make information accessible

8. Why to study DS? (3) Learn to enable distant communication
, chat, skype

9. Why to study DS? (4)
Learn to solve tough problems

10. Goals
Understand principles and concepts underlying computer networks and distributed systems
Layering, communication, coordination
Learn how to design reliable, cooperative (basic) distributed applications
Gain practical skills on the development of simple distributed systems (in Java)
Improve your software engineering expertise

11. Areas Distributed systems hide the network complexity …
… but in order to understand and develop them, we need competencies in
Computer networks
Concurrency & Inter-process communication
Distributed
Systems
Computer
Networks
Concurrency
&
Inter-process
Communication

12. Course Structure 24 lectures (some will be very practical)
12 programming labs in 2 groups
5 of these are with Raspberry PIs
5 Assignments

13. Tentative Schedule Introduction RMI OSI Overview RMI (Lab)
Physical Layer
(Guest lecture)
Data Link Layer
Bobo (Guest lecture)
Link Layer (P&P Lab)
RPI 2: Distributed Algorithms (Lab)
Link Layer Simulation (Lab)
P2P
Medium Access Control Layer
P2P (continued)
Network Layer Theory
RPI 3: Distributed MD5 Cracker (Lab)
Wireshark I (Lab)
P2P (Lab)
Network Layer on the Internet
Cryptography
RPI 1: Networking basics (Lab)
RPI 4: Distributed MD5 Cracker (Lab)
Transport Layer
RPI 5: Distributed MD5 Cracker (Lab)
Transport Layer/Java Sockets
Cryptography (P&P Lab)
Java Sockets (Lab)
Multiagents+AntMe
Application Layer
AntMe (Lab)
Wireshark II (Lab)
Coordination
Conceptual Exercises (P&P Lab)
AntMe competition
Exam preparation

14. Software Used languages/tools/environments:
Java
C# (MS Visual Studio 2012 or later)
Wireshark
Raspberry PIs provided by university
Software is available on the virtual machines provided by the faculty

15. Assignments Hamming Codes Java Socket Chat Java RMI Chat
Distributed MD5 Cracker on RPIs
Multiagent programming
You can work in teams of 2
First three assignments count 5% each, last two 10% each towards the final grade ( total max. 35%).
Only assignments that are better than the exam grade are taken into account ( Assignments can only improve grade)
Assignments are optional

16. Assignments are optional… Should I do them?
All but one student submitting less than 3 assignments failed
All students submitting at least 3 assignments passed

17. Plagiarism
Plagiarism  0 points for all assignments + notification of the study course leader
What is plagiarism?
Any copying of work of others (other students, other people on the web, yourself from last year)
Minor copying is only ok, if clearly indicated
What is worse than plagiarism?
Attempting to conceal plagiarism
Specialized software is used….

18. Exam Allowed to bring one handwritten A4 cheatsheet (both sides)
Develop it yourself and magic will happen: The things you put there, you won’t need to look up
Dictionaries
Pens
Nothing else (no calculators, no textbooks, …)
Anything covered in the lectures and labs can be on the exam
No coding questions
But modelling/design questions are possible
You do not need to know everything (achieving ~90% of the points is enough to get grade 30)
Some exams from previous years are found online

20. Grading Exam: 65% Five assignments: Up to 35%
Only the assignments with a grade higher than the exam mark count
 Assignments only improve the overall grade
To pass the course, the exam grade has to be at least 18
Assignments are valid for all three exam sessions

21. Grading (2) Final grade = (1-AW)*E + AW*AG*((E+100)/(118))
AW = weight of assignments that are at least as good as the exam grade ( )
AG = average grade of the assignments that are at least as good as the exam grade
E = exam grade

22. Material The first half of the course is covered in
A.S. Tanenbaum. Computer Networks. Prentice Hall (there are also 5 copies in German in the library)
J.F. Kurose, K.W. Ross. Computer Networking – A Top-Down Approach (5th edition). Pearson Education
For the second half, the best resource are the slides and online resources
Some content regarding RPC, P2P, Cryptography and Coordination is in
A.S. Tanenbaum, M. van Steen. Distributed Systems: Principles and Paradigms. Prentice Hall.
G. Coulouris, J. Dollimore, T. Kindberg. Distributed Systems: Concepts and Design. Addison-Wesley.

23. Acknowledgment Part of the slides are based on: Slides by Tanenbaum
Previous versions of this course by Nutt, Montali and Pirro

24. Questions?

25. Introduction to Distributed Systems
What is a distributed system?
History
Paradigms

26. Name some distributed systems
Collect examples

27. What is a Distributed System?
Ask for proposals for definition
If you ask three computer scientists, you will get four different opinions
RPI power an fragen ob DS ist?

28. Distributed System
Some definition: A collection of independent computers that appears to its users as a single coherent system
 Several computers
Independence
Heterogeneity of architectures
 Provides service
 Transparency (users perceive a single system)
Collaboration!
Realizing collaboration mechanisms is the goal of distributed systems development
Interaction is hidden

29. Alternative Definition
A distributed system is one where a machine I've never heard of can cause my program to fail
Leslie Lamport
Abstraction: From the outside, a black box providing service interfaces

30. Requirements
A collection of independent computers that appears to its users as a single coherent system
Make resources/services accessible
Hide complexity (Transparency)
Openness (wrt. extensions and heterogeneous components)
Scalability (wrt. performance)

31. Requirement 1 Make Resources/Services Accessible
Resource: source or supply from which a benefit is produced
printers, computers, files, pages, networks, project deliverables, notes, …
Services:
Computations, data enrichment, ..
Issue: security
Balance between sharing and privacy
Difficult to be achieved in “open” networks

32. Requirement 2 Transparency
Distributed system should present itself to users and applications as a single coherent system
Transparency
Description
Access
Hide differences in data representation and how a resource is accessed
(multiple interacting OSs)
Location
Hide where a resource is located (naming strategies)
Migration
Hide that a resource may move to another location
Relocation
Migration while resource is in use
Replication
Hide that a resource is replicated (requires location transparency)
Concurrency
Hide that a resource may be shared by several competitive users
Failure
Hide the failure and recovery of a resource
Persistence
Hide whether a (software) resource is in memory or on disk
Whatsapp both on Android and Iphone
Concurrency: Printer no, webserver yes
Replication and concurrency: classical issues for databases

33. Requirement 2 Transparency
Sometimes transparency is impossible
Performance
Computation/interaction timings (e.g. Skype)
Concurrency issues (e.g., Dropbox)
Location-awareness (embedded/ubiquitous systems)
Comprehensibility
Network awareness clarifies “strange behaviors”
 Properly tune the degree of transparency
Distributed system should present itself to users and applications as a single coherent system

34. Requirement 3 Openness
Open DS: offers services according to standard rules describing their syntax and semantics
Interoperability: extent by which two systems/components can work together by just knowing each other’s interfaces
Portability: to what extent an application can be seamlessly executed in a different environment
Reconfiguration/Extensibility: to what extent the DS can be re-organized or accommodate new-parts
Skype vs bitcoin/bittorrent – bittorrent can write your own client!

35. Standardization Important for Standardization efforts: Sustainability
Extendibility
Security
Standardization efforts:
ISO = International Standards Organisation
ITU-T = International Telecommunication Union
IETF = Internet Engineering Task force
IEEE = Institute of Electrical and Electronic Engineers
W3C = World Wide Web Consortium

36. Importance of Standards
Guarantee large-scale interoperation
Guarantee heterogeneity of products
Network standards: regulate interaction between remote hardware/software entities, connected through a network
Two types of standard
De jure (top-down): international committees and organizations
De facto (bottom-up): emerge from practice (UNIX, JAVA, TCP/IP, …)
Heterogeneity: Android
Draw timing

37. Openness Guidelines High componentization
Small replaceable/adaptable components with clear interfaces
Separation of logic and implementation
Specify only the logic
Leave implementation to the developers

38. Sample from the HTTP Protocol Specification
Negotiation An HTTP/1.1 server MAY assume that a HTTP/1.1 client intends to maintain a persistent connection unless a Connection header including the connection-token "close" was sent in the request. If the server chooses to close the connection immediately after sending the response, it SHOULD send a Connection header including the connection-token close. An HTTP/1.1 client MAY expect a connection to remain open, but would decide to keep it open based on whether the response from a server contains a Connection header with the connection- token close. In case the client does not want to maintain a connection for more than that request, it SHOULD send a Connection header including the connection-token close. If either the client or the server sends the close token in the Connection header, that request becomes the last one for the connection. Clients and servers SHOULD NOT assume that a persistent connection is maintained for HTTP versions less than 1.1 unless it is explicitly signaled.

39. Requirement 4 Scalability
Scalable DS: operates effectively and efficiently independently from the number of resources and users
Dimensions of scalability (Neumann, 1994)
Size: w.r.t. number resources/users
Geography: w.r.t. location of resources/users
Administration: w.r.t. involved administrative silos

40. Scalability Problems (Bottlenecks)
Concept
Example
Centralized services
A single server for all users
(legacy systems, security reasons)
Centralized data
A single database
(avoiding synchronization issues)
Centralized algorithms
Doing routing based on complete information
Decentralized algorithms
No machine holds the complete system state
Machines take decisions only on local info
The algorithm is robust to machine failure
Routing means network routing – like IP

41. Scalability Problems (Geography)
From LANs to WANs…
LAN: short distances
WAN: large distances
LAN
WAN
Synchronous communication
(wait for answer)
YES NO
Broadcasting
(send to all)
Impossible
Centralized solution
MAYBE
Broadcasting: needed for locating a service – in WAN not possible, need other techniques

42. Scalability Problems (Admin.)
Combination of two administrative domains…
Non-technical issues
Politics, human relationships…
Conflicting policies
Payment, management, security
Lack of mutual trust
Internal code: extensively tested/used
External code: unknown
Security issues
Protection from malicious attacks coming from
The new users that have access to the system
The foreign code that is now part of the system

43. Scaling Techniques Reduction of the the overall communication
Thin clients vs. fat clients
2. Caching
Consistency and synchronization issues
3. Increase of computational power
Works only for the size-related aspects
4. Increase of number of computational resources
Temporary (cloud solutions)
Works only if task can be well parallelized

44. False Myths The network is reliable The network is secure
The network is homogeneous
The topology does not change
Latency is zero
Bandwidth is infinite
Transport cost is zero
There is one administrator
We deal in the lecture what can be done against many of these issues

45. Issues in Distributed Systems
Concurrency
Multiple autonomous processes running in parallel
Cooperation (resource sharing)
Competitiveness (shared access)
No global clock
Every interacting process has its own local time
No synchronization
Independent failures
Machine crash
Network problems
Security

46. How about our examples? Make resources/services accessible
Transparency
Access
Location
Migration
Relocation
Replication
Concurrency
Failure
Openness
Interoperability
Portability
Extensions
Scalability
Size
Geography
Administration

47. 2. History

48. Computers 1945 – mid 80s Large-sized, expensive computers
Centralized computing (single-processor systems)
Monolithic programming
OS: monolithic kernel
Imitation Game

49. Computers around 1985 Powerful micro-processors
Moore’s law: the number of transistors that can be placed over an integrated circuit doubles every 2 years
Computer (local-area) networks
Connection among several machines in a building
Small amount of data quickly transferable (μs)
Micro-kernels
Functionalities moved in user space
LAN  functionalities distributed across the network!

50. Centralized vs. Distributed Computing

51. Computers 1985-now High-speed LANs Wide area networks Internet
100 Mbps – 10 Gbps
Wide area networks
Connection at the earth level
64 Kbps – 1 Gbps
Internet
Network of networks
Billions of users

52. Distributed Systems vs Computer Networks
Distributed System: transparency/coherence
Middleware manages interaction (black box)
Applications often run on top of a computer network
E.g. WWW, multiplayer online games, …
Computer network: collection of autonomous computers interconnected by a single technology
Interconnection = ability of exchanging information
Usually via message passing
Users exposed to the actual machines (white box)
Applications support users in interconnecting machines
E.g. remote desktop, file transfer, remote printers, …
To realize distributed systems we need to understand/program computer networks!

53. The Internet A network of networks built upon TCP/IP
A basis for distributed systems (www, name services, VoIP, file sharing, …)

54. How Many People? http://www.internetlivestats.com/internet-users/
View them all, then click on one to see the colors

55. How Big is the Internet?
“Eric Schmidt, the CEO of Google, the world’s largest index of the Internet, estimated the size at roughly 5 million terabytes of data. That’s over 5 billion gigabytes of data, or 5 trillion megabytes. Schmidt further noted that in its seven years of operations, Google has indexed roughly 200 terabytes of that, or .004% of the total size.”
They index only first part of large webpages!

56. Network Applications Finance and commerce
eCommerce e.g. Amazon and eBay, PayPal, online banking and trading , Bitcoin
The information society
Web information and search engines, ebooks, Wikipedia; social networking: Facebook, Google+.
Creative industries and entertainment
online gaming, music and film at home, user-generated content, e.g. YouTube, Flickr
Healthcare
health informatics, online patient records, monitoring patients
Education
e-learning, virtual learning environments; distance learning
Transport and logistics
GPS in route finding systems, map services: Google Maps, Google Earth
Science
Access to research papers and data sets, collaborative working (SVN, Google Docs, Overleaf), Mapreduce,
Environmental management
sensor technology to monitor earthquakes, floods or tsunamis

57. Learned today What is a distributed system Requirements
Service provision with distributed black-box execution
Requirements
Accessibility
Transparency
Openness
Scalability
Examples of DS and level of satisfaction of the requirements
Next: Protocol Layers