Chap 01 Updated

8/3/2019 Chap 01 Updated

1/34

Tanenbaum & Van Steen, Distributed Systems: Principles and Paradigms, 2e, (c) 2007 Prentice-Hall, Inc. All rights reserved. 0-13-239227-5

DISTRIBUTED SYSTEMS

Principles and ParadigmsSecond Edition

ANDREW S. TANENBAUMMAARTEN VAN STEEN

Chapter 1Introduction


2/34


Definition of a Distributed System (1)

A distributed system is:

A collection of independentcomputers that appears to its

users as a single coherentsystem.


3/34


Definition of a Distributed System (2)

Figure 1-1. A distributed system organized as middleware. Themiddleware layer extends over multiple machines, and offers

each application the same interface.


4/34

Goals of Distributed Systems To make resources easily accessible

Share printers, supercomputers, storage

Transparent ( Hide the fact that resources are distributedacross a network)

Open

Interoperability, Portability, extensible

Scalable Size, location and Administration



5/34

Problems

Design difficult:

Operating systems, programming languages, administration

Security Network: Disconnect, loss of data



6/34


Transparency in a Distributed System

Figure 1-2. Different forms of transparency in adistributed system (ISO, 1995).


7/34

Scalability

Size: more users & resources

Geographic location

Administration



8/34Tanenbaum & Van Steen, Distributed Systems: Principles and Paradigms, 2e, (c) 2007 Prentice-Hall, Inc. All rights reserved. 0-13-239227-5

Scalability Problems

Figure 1-3. Examples of scalability limitations.


9/34

Scaling Techniques

Three techniques for scaling:

Hiding Communications latencies

Avoid waiting as much as possible (Asynchronous communication)

Distribution Splitting a component of the system into smaller parts, and spreading across

the system e. g. (DNS), WWW

Replication

Replicate components across the system Caching (copy of a resource)

Consistency problems (how much to tolerate?)




Scaling Techniques (1)

Figure 1-4. The difference between letting (a) a serveror (b) a client check forms as they are being filled.



Scaling Techniques (2)

Figure 1-5. An example of dividing the DNS

name space into zones.



Pitfalls when DevelopingDistributed Systems

False assumptions made by first time developer:

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.


13/34

Types of Distributed Systems

1. Distributed Computing Systems

2. Distributed Information Systems3. Distributed Pervasive Systems



14/34

Distributed Computing Systems

Distributed Computing Systems

Cluster computing

Homogeneity (same OS & network)

E.g. Linux-based Beowulf

Grid Computing

Heterogeneity (no assumptions)

Resources from different organizations are brought together

to allow the collaboration of a group of people or institutions




Cluster Computing Systems

Figure 1-6. An example of a cluster computing system.


16/34

Grid ComputingA layered architecture of Grid Computing:

Fabric layer: provides interfaces to local resources, to allowsharing of resources

Connectivity layer: Consists of Communication & security

protocols to support Grid transactions

Resource Layer: manage a single resource, perform specificoperation

Collective layer: Handling access to multiple resources, e.g.

resource discovery, allocation & scheduling of tasks

Application Layer: consists of applications that operate withina virtual organization




Grid Computing Systems

Figure 1-7. A layered architecture for grid computing systems.


18/34

Distributed Information Systems

Distributed Information Systems

Transaction Processing Systems

Enterprise Application Integration




Transaction Processing Systems (1)

Figure 1-8. Example primitives for transactions.




Characteristic properties of transactions: Atomic: To the outside world, the transaction

happens indivisibly.

Consistent: The transaction does not violatesystem invariants.

Isolated: Concurrent transactions do notinterfere with each other.

Durable: Once a transaction commits, thechanges are permanent.


21/34



Figure 1-9. A nested transaction.


22/34



Figure 1-10. The role of a TP monitor in distributed systems.


23/34


Enterprise Application Integration

Figure 1-11. Middleware as a communication facilitator inenterprise application integration.


24/34

Enterprise Application IntegrationTypes of communication Middleware:

Remote procedure call (RPC)

Remote method invocation (RMI)

Message oriented Middleware (MOM)



25/34

Distributed Pervasive Systems

Default is instability

Small, battery-powered, mobile, wireless connection

Lack of human administrative control



26/34


Distributed Pervasive Systems

Requirements for pervasive systems

Embrace contextual changes.

Encourage ad hoc composition. Recognize sharing as the default.


27/34


Home systems

Systems built around home networks

E.g. TV, Video, PDA, Kitchen appliances

Should be self-configuring & self-managin


28/34


Electronic Health Care Systems (1)

Questions to be addressed for health care systems: Where and how should monitored data bestored?

How can we prevent loss of crucial data?

What infrastructure is needed to generate andpropagate alerts?

How can physicians provide online feedback?

How can extreme robustness of the monitoringsystem be realized?

What are the security issues and how can theproper policies be enforced?


29/34


Electronic Health Care Systems (2)

Figure 1-12. Monitoring a person in a pervasive electronic healthcare system, using (a) a local hub or

(b) a continuous wireless connection.


30/34


Sensor Networks (1)

Questions concerning sensor networks:

How do we (dynamically) set up anefficient tree in a sensor network?

How does aggregation of results takeplace? Can it be controlled?

What happens when network links fail?


31/34


Sensor Networks (2)

Figure 1-13. Organizing a sensor network database, while storingand processing data (a) only at the operators site or


32/34


Sensor Networks (3)

Figure 1-13. Organizing a sensor network database, while storingand processing data or (b) only at the sensors.


33/34

Summary

Distributed Systems (DS) consist of autonomous computersthat work together to give the appearance of a singlecoherent system

DS make it easier to integrate different applications runningon different computers into a single system

Well designed DS scale well with respect to the size of theunderlying network.

Cost: more complex software, performance & security



34/34

Summary DS aim to hide distribution of processes, data & control

(distribution transparency)

Many developers make wrong assumptions when developing

DS Different types of DS

Emerging class of DS, ubiquitous computing (not managedthrough system administrator)

Documents

Chap 01 Updated