CdLM Ingegneria Informatica

(Politecnico di Bari)

A.A. 2016-17

Distributed systems

Module 11 > Distributed System Architectures (part D)

Prof. S.Pizzutilo

Distributed Systems
A Distributed System is a collection of autonomous processors :
•  interconnected by a communication network,
•  with own resources (processing capabilities, local memories, local
data and softwares ),
•  with the ability to coordinate their activities and to share the
everywhere located resources.

Distributed System Architecture = number, type and interaction of

hardware/software components

1
 Distributed Systems types

Distributed computing systems:

•  Cluster
•  Grid
•  Cloud (?) Information systems:
•  Transactional Systems
•  Integrated company applications

Pervasive systems:
•  Home automation systems
•  Distributed systems for health care
•  Sensors network
•  Multiagent systems

Distributed Information Systems

Information systems realised to allow the integration

(interoperability) between different applications.

IT systems are divided into transactional (OLTP) and analytical

(OLAP).

OLTP (On-line Transaction Processing) systems provide source

data to data warehouses, whereas
OLAP (On-line Analytical Processing) systems help to analyze
it.

2
 OLTP is characterized by a large number of short on-line transactions
(INSERT, UPDATE, DELETE). The main emphasis for OLTP systems is put on
very fast query processing, maintaining data integrity in multi-access
environments and an effectiveness measured by number of transactions per
second.

OLAP is characterized by relatively low volume of transactions. Queries are often

very complex and involve aggregations. For OLAP systems a response time is an
effectiveness measure. OLAP applications are widely used by Data Mining techniques. In
OLAP database, there is aggregated, historical data, stored in multi-dimensional schemas.

OLTP System OLAP System

Online Transaction Processing Online Analytical Processing
(Operational System) (Data Warehouse)
Operational data; OLTPs are the original source Consolidation data; OLAP data comes from the
Source of data
of the data. various OLTP Databases
To help with planning, problem solving, and
Purpose of data To control and run fundamental business tasks
decision support
Reveals a snapshot of ongoing business Multi-dimensional views of various kinds of
What the data
processes business activities
Short and fast inserts and updates initiated by
Inserts and Updates Periodic long-running batch jobs refresh the data
end users
Relatively standardized and simple queries
Queries Often complex queries involving aggregations
Returning relatively few records
Depends on the amount of data involved; batch
data refreshes and complex queries may take
Processing Speed Typically very fast
many hours; query speed can be improved by
creating indexes
Larger due to the existence of aggregation
Can be relatively small if historical data is
Space Requirements structures and history data; requires more indexes
archived
than OLTP
Typically de-normalized with fewer tables; use of
Database Design Highly normalized with many tables
star and/or snowflake schemas

Backup religiously; operational data is critical Instead of regular backups, some environments
Backup and Recovery to run the business, data loss is likely to entail may consider simply reloading the OLTP data as a
significant monetary loss and legal liability recovery method

source: www.rainmakerworks.com

3
 Transactional systems (1)

Transaction processing systems consist of computer hardware and software

hosting a transaction-oriented application that performs transaction
routines necessary to conduct business.

Four critical properties (ACID) of transactions :

Ø  Atomicity: The transaction executes completely or not at all. The successful
completion of a transaction is called commit. The failure of a transaction is called abort..
Ø  Consistency: The transaction preserves the internal consistency of the
database. The database at least satisfies all its integrity constraints..
Ø  Isolation: The transaction executes as if it were running alone, with no other
transactions. An execution is serializable (meaning isolated) if its effect is the same as
running the transactions serially,, in sequence, with no overlap in executing any two of them.
Ø  Durability: The transaction's results will not be lost in a failure. When a
transaction completes executing, all its updates are stored in a persistent storage.

Transactional Systems (2)

TS Components:
• End-user device: is someone who requests the execution of transactions, such as a customer of a bank
or of an Internet retailer. An end-user device could be a physical device, such as a cash register or gasoline
pump.

• Front-end program: is an application code that interacts with the end-user device. Usually it sends and
receives menus and forms, to offer the user a selection of transactions to run and to collect the user's input.
Often, the device is a web browser and the front-end program is an application managed by a web server
that communicates with the browser via HTTP.

• Request controller: is responsible for receiving messages from front-end programs and turning each
message into one or more calls to the proper transaction programs. In a distributed TP system, it requires
sending the message to a system where the program exists and can execute.

• Transaction server: is a process that runs the parts of the transaction program that perform the work the
user requested, typically by reading and writing a shared database, possibly calling other programs, and
possibly returning a reply that is routed back to the device that provided the input for the request.

• Database system: manages shared data that is needed by the application to do its job.

4
 Transactional System Software (3)

o  Transactional middleware = a layer of software components between TP

applications and lower level components such as the operating system,
database system, ….

o  A SOA-based TP system may be assembled using a combination of services

from a variety of applications and using a variety of operating systems,
middleware platforms, and programming languages.

o  A TP application may exist as a combination of reusable services.

o  Representational State Transfer (REST) is another approach to SOA, rather

different than that of Web Services. The term REST is used to denote the
protocol infrastructure used for the World Wide Web (HTTP); and to denote a
software architectural pattern that can be implemented by web protocols
(REST/HTTP).

Integrated company applications

Enterprise application integration is an integration framework

composed of a collection of technologies and services which
form a "middleware framework" to enable integration of
systems and applications across an enterprise.
Applications for managing customers, business intelligence applications,
payroll and human resources systems typically cannot communicate with one
another in order to share data or business rules.

An ERP covers many functions:

•  customer relationship management (CRM), dealt directly with customers,
•  e-business systems such as e-commerce, e-government, e-telecom, and e-
finance,
•  supplier relationship management (SRM) became integrated later, when the
Internet simplified communicating with external parties.

5
 ERP
The ERP is the integrated management of core business processes, often in real-
time and mediated by software and technology. It provides an integrated and
continuously updated view of core business processes using common databases
maintained by a database management system.
Technical solutions for the Enterprise application integration include (a) rewriting part of the
delivered software, (b) writing a homegrown module to work within the ERP system, or (c)
interfacing to an external system. These three options constitute varying degrees of system
customization, with the first being the most invasive and costly to maintain.

An ERP system typically includes the following characteristics and modules:

•  Operates in real time,
•  Requires a transactional database that supports all the applications,
•  Requires a management portal ( a dashboard),
•  Supplies chain management: chain planning, supplier scheduling, product
configurator, order to cash, purchasing, inventory, warehousing,….
• A consistent look and feel across modules,
• Customer relationship management.

ERP critical points

•  Customization is always optional, whereas the software must always be

configured before use.
•  The software is designed to handle various configurations, and behaves
predictably in any allowed configuration.
•  The effect of configuration changes on system behavior and performance is
predictable and is the responsibility of the ERP vendor. The effect of
customization is less predictable.
•  Configuration changes survive upgrades to new software versions. Some
customizations survive upgrades, though they require retesting. Other
customizations (e.g., those involving changes to fundamental data structures)
are overwritten during upgrades and must be re-implemented.

6
 Pervasive Distributed Systems

They are unstable distributed systems in the HW/SW

structure and power, in the communication forms and in their
position, built specifically to resolve problems of ubiquitous
computing , disappearing computing, mobile computing, and
embedded computing.

• Computing becomes pervasive or ubiquitous,

• Rise of “devices”,
• Computing everywhere,
– smart cities, smart homes, smart highways, smart
classroom, ...

Pervasive Distributed Systems

Distributed systems with mobile devices that generally communicate using ad

hoc wireless protocols to realise specific computational objectives.

7
 Universitat Linz - Institut fur Pervasive Computing

Rise of Pervasive Computing

Ø  Household and personal systems: PDA, smart phone,

recommender systems, …..
Ø  Personal health monitoring: Sensors to monitor fitness,
diabetes, blood pressure, detect falls
Ø  Sensor networks: Mesh network (wireless network of sensors or
low computing capability with at most 2 communication paths between
each node) , MANET (Mobile Ad hoc NETwork), ….

Ø  Driven by miniaturization of computing

Tiny sensors with computing and communication capability
Ø  Internet of things
Devices with the ability to communicate and to cooperate over Internet

8
 Technologies for pervasive computing

•  Wireless communication protocols

•  Intelligent Software Agents
•  Intelligent interfaces
•  Smart devices
•  Awareness
•  Distributed software

WI-FI architectures

Wireless Protocols: RFID, DECT, IrDA, Bluetooth,

IEEE802.11
§  Infrastructural
§  Ad Hoc §  More cells or devices (compute
§  Wireless devices nodes + Access Point = called Basic
Service Set = BSS )
§  No access points to the wired
§  Each cell is controlled by an AP
network ( IBSS= Independent
(Access Point)
Basic Service Set)

9
 A MANET example

Software Agent Properties

(Wooldridge & Jennings 1995)

A weak notion of Agent = computational software system with :

Ø  autonomy: agents operate without the direct intervention of humans or others,
and have some kind of control over their actions and internal state;
Ø  social ability: agents interact with other agents (and possibly humans) via
some kind of agent-communication language
[Genesereth and Ketchpel, 1994];
Ø  reactivity: agents perceive their environment, (which may be the physical
world, a user via a graphical user interface, a collection of other agents, the
INTERNET, or perhaps all of these combined), and respond in a timely
fashion to changes that occur in it;
Ø  pro-activeness: agents do not simply act in response to their environment,
they are able to exhibit goal-directed behaviour by taking the initiative.

Ref: B.De Carolis “Sistemi ad Agenti” CdLM Informatica UNIBA 2003

10
 A stronger notion of Agent
Agent conceptualised or implemented using concepts that are more usually
applied to humans

q  Agent using mentalistic notions, such as knowledge, belief, intention, and
obligation [Shoham, 1993].
q  Emotional agents [Bates, 1994][Bates et al., 1992°] = designing and building
agents in terms of human-like mental states .
q  Agents visually represented , by using a cartoon-like graphical icon or an
animated face [Maes, 1994a] –

• mobility = the ability of an agent to move around an electronic network [White,

1994];
•  veracity = the assumption that an agent will not knowingly communicate false
information [Galliers, 1988];
• benevolence = the assumption that agents do not have conflicting goals,
[Rosenschein and Genesereth, 1985];
• rationality = (crudely) the assumption that an agent will act in order to achieve
its goals, and will not act in such a way as to prevent its goals being achieved -
at least insofar as its beliefs permit [Galliers, 1988].

Intelligent Agents
Software entity with :
ü  Ambient perceptors (sensors)
ü  Capability to execute actions in an enviornment (effectors).
[Russel & Norvig]

Functional properties

A rational agent acts so that to maximize its performance

efficiency
Ref: B.De Carolis “Sistemi ad Agenti” CdLM Informatica UNIBA 2003

11
 Agent types

q  “intelligent” Agents (Artificial Intelligence)

q  Distributed intelligent Agents (Distributed AI)
q  Interface Agents (Human-Computer Interaction)
q  MAS = Multi Agent Systems
q  Conversational Agents
AI HCI

DAI
ref;: B.De Carolis “Sistemi ad Agenti” CdLM Informatica UNIBA 2003

Agent Classification

12
 Software Agent Types

Agent type   Sensing   Actions   Objectives   Environment  

Symptoms,
questions, patient well-
health diagnosis medical patient,
assessment, being, minimize
systems   verdict, patient hospital  
therapy   costs  
reaction  
pixels of
satellite images variable print a scene correct the satellite
analisys   intensity and classification   classification   images  
colour  
pixels of
pick-up and position objects conveyor
picker robots   variable
organize objects   in a correct box   band  
intensity  
open and close
temperature maximize
valve,
refinery control   and pression product clarity refinery  
temperature
reading   and security  
control  
interactive print exercises,
maximize
english language words typed   text amendments classroom  
student scores  
tutor   and suggestions  

New paradigms of the Man-Machine Interaction

Ubiquitous computing : “a possible next generation computing

environment in which each person is continually interacting with
hundreds of nearby wireless interconnected computers” (Weiser,
Xerox 1980)

New concepts :
•  Mobile New interfaces:
•  nomadic •  Intuitive
•  Pervasive •  Seamless interfaces
•  disappearing / invisible •  Tangible artefacts
•  Wearable computers •  Haptic devices
•  Embedded interactive objects

Ref: . De Rosis “Interazione uomo-macchina” CdLT Informatica e Com.Digit. UNIBA 2004-05

13
 New Paradigm basics

ü  Declining of individual relationshp between user and device.

ü  Distributing computation capability on different devices,
ü  Simultaneous access of different users to the same device.
ü  Integration between computation and observation users devices,
ü  Network connection between computation devices and users,
ü  Automatic survey of user actions,
ü  Independence of computational devices and its initiative
capability.

Ref: F. De Rosis “Interazione uomo-macchina” CdLT Informatica e Com.Digit. UNIBA 2004-05

Multi Agent System

14
 MAS

Coordination and Negotiation strategies

Multi agent planning

Coordination and Negotiation

distributed strategies

A conversational multiagent system

I‘ve been eating too much
for about one year and I
realise that I’m not myself,
that I’m not thinking quite
Emotional Intelligence:
right!
- Shallow: ability to exhibit an
Oh, I’m sorry,
that’s a bad
emotional state
situation
indeed...
- Inner: emotions drive
reasoning behind the dialog.
Believable ECA providing
advice in domains influenced how the dialog may be
by affective factors: affected by the emotional state
i.e. eating disorders of the two interlocutors and by
their personalities.

15
 A conversational multiagent system

AA.VV. “Towards a Socially Intelligent ECA” CHItaly 09

Wearable Computers
(from MIT Media Lab)

To date, …most machines sit on the desk and interact with

their owners for only a small fraction of the day.
Smaller and faster notebook computers have made mobility
less of an issue, but the same staid user paradigm persists.

Wearable computing hopes to shatter this myth of how a

computer should be used.
«A person’s computer should be worn, much as eyeglasses
or clothing are worn, and interact with the user based on the
context of the situation.»

With heads-up displays, unobtrusive input devices,

personal wireless local area networks and a host of other
sensing and communication tools, the wearable computer
can act as an intelligent assistant.

Ref: F. De Rosis “Interazione uomo-macchina” CdLT Informatica e Com.Digit. UNIBA 2004-05

16
 References
•  A.Tanenbaum, M.Van Steen “Distributed systems” Pearson Prentice Hall 2007 ) cap. 2
•  L.Manelli “ Abstract State Machine per Modellare Servizi di Grid Computing” tesi di
dottorato in Informatica A.A.2012-13
•  B.De Carolis “Sistemi ad Agenti” lucidi corso per “Informatica” magistrale Uniba 2003
•  F. de Rosis “Interazione Uomo-Macchina” lucidi corso per CdLT in ICD Uniba A.A.
2004-2005

17