More Software Architectures
Blackboard Architecture
• In a blackboard system, a set of problem solving modules
(typically called knowledge sources) share a common global
database (called the blackboard).
• Knowledge sources respond to changes on the blackboard,
and interrogate and subsequently directly modify the
blackboard
Blackboard Architecture
• To solve a complex problem, the intelligent agents cooperate
as functional specialists, observing updates to the blackboard
and self-actualizing (in an event driven process) when there is
new information to process.
• Agents continually update the blackboard with partial
solutions when the agents capabilities for processing match
the state of the blackboard.
Blackboard Architecture
• The blackboard architecture is a distributed computing model
for a metaphor describing how people work together to
collaboratively solve a problem around a blackboard
(whiteboard in todays lingo).
Blackboard Architecture
• The blackboard architecture can be very effective in solving
complex distributed computing problems, including event
processing problems; however, scheduling the self-actuating
agents can be a key challenge.
References
• A Brief Introduction to Blackboard Architectures. (2007, 08 20).
Retrieved 01 31, 2012, from http://www.thecepblog.com:
http://www.thecepblog.com/2008/07/20/a-briefintroduction-to-blackboard-architectures/
• Hunt, :. D. (2002, 05 27). Blackboard Architectures. Retrieved
01 31, 2012, from http://www.agent.ai:
http://www.agent.ai/doc/upload/200402/hunt02_1.pdf
SOFTWARE ARCHITECTURES:
INTERPRETER
TEAM 2
FEBRUARY 2, 2012
INTERPRETER
•
Also called virtual machine
•
Parses and executes input commands, updating
the state maintained by the interpreter
•
Basic Idea:
•
A program is written in a particular language, the
interpreter translates the language, and begins to act
upon it
COMPOSITIONS
• Interpreter Engine
• Finishes work of interpreter
• Data Store Field
• Includes source code
• Interpreter State
• Data structure that stores state of the engine
• Program State
• Records progress of interpreted code
INTERPRETER
DISADVANTAGES
• Hard to design and test
• Cannot generate all possible programs that could
possibly be executed
ADVANTAGES
• Ideal for end-user programmability
• Permits flexibility
• Supports dynamically changing the set of
capabilities
• Highly dynamic behavior allowed
• Can combine with other architectural styles
QUESTIONS?
REFERENCES
• Software Architecture:
http://books.google.com/books?id=WHMksQOj6ScC&pg=PA76&lpg=PA76&dq=interpret
er+software+architecture&source=bl&ots=Wz3582xfgT&sig=E18v3bzd0LMGcKGHQynBW5
Y5dh0&hl=en&sa=X&ei=waMoT4CBCuiU2AWOy83iAg&ved=0CHcQ6AEwCA#v=onepage
&q=interpreter%20software%20architecture&f=false
• Interpreted Program:
http://www.thomasalspaugh.org/pub/fnd/architecture.html#Interpreter
• Architectural Styles Software Lecture:
http://www.google.com/url?sa=t&rct=j&q&esrc=s&source=web&cd=1&ved=0CCgQFjAA
&url=http%3A%2F%2Fsunset.usc.edu%2Fclasses%2Fcs578_2012b%2F05_Architectural_Styles
.ppt&ei=9HkoT8u4GoSc2AWvvNneAg&usg=AFQjCNEJT5OYN9cKGFTbs8apyMPYN66RlQ
Team 3
Plug-in Architecture
http://prezi.com/esg2ox3qu9rz/plug-in-softwarearchitecture/
SEARCH ORIENTED SOFTWARE ARCHITECTURE
Search engine is the main integration component of the architectural layer
 As opposed to the traditional relational database management system
 Keyboard search instead of SQL queries
Overview of Search-Oriented Design
In Depth Look
SEARCH ORIENTED SOFTWARE ARCHITECTURE
Benefit: rapid response time and more convenient, in terms as SQL versus keyboard search
Drawback: The data tier is not as organized
REFERENCES
Information:
• courses.utep.edu/portals/870/SoftwareArchitecture
• http://www.mediabuzz.com.sg/asian-emarketing/november2011/1424-search-oriented-architecture
Images:
• http://www.soainstitute.org/images/contributors/Paradkar_March11_
2_Figure2.jpg
• http://www.soainstitute.org/images/contributors/Paradkar_March11_
2_Figure1.jpg
Thank You
Space – Based Architecture
Team 5
Distributive Programing
• Remote Procedure Calls (RPC)
– Objects communicate and synchronize their
process through procedure calls. CORBA, EJB, and
COM.
– The greater number of participating objects, the
longer and more conversational paths become.
• Space-Based Architecture simplifies this idea
Space – Based Architecture
(Blackboard Communication)
• In a space implementation, the space itself
synchronizes all of its participants. This is called
Blackboard communication.
• In Blackboard communication, Participants:
– Know nothing about one another (anonymous
exchange)
– Do not need to be using the same process or machine
– Independent of one another
Distributed Caching
• It is often difficult, by weighing costs and
benefits, to achieve an optimal load balance in
a system.
• Since workers are well informed about the
load upon their respective machines, a “space
proxy” is charged with request redistribution.
Only requests for which a worker with
sufficient resources can be found will be
accepted.
Conclusion
• Advantages of a space-based approach:
– Faster development cycles
– High scalability by the addition of cheap standard
hardware
– Shift of load from web server to concurrent
worker processes.
– Total decoupling of interfaces
– Distributed Caching
Software Architecture Design:
Share Nothing
• Related to the multiprocessor high transaction rate
family of designs:
Shared Memory, Shared Disks, Shared Nothing
• Like other multiprocessor architecture designs, shared
nothing is a distributed computing architecture.
• A distributed system is made up of multiple computers
that communicate through a computer network. The
computers interact with each other in order to achieve
a common goal.
Shared Nothing (cont.)
• Shared nothing systems, like the name implies, do not
share memory nor disks among the computers within,
called nodes.
• Instead each node is independent and self-sufficient
and does not depend on any outside resource.
Shared Nothing
Shared Everything
Reference: http://gerardnico.com/wiki/data_storage/shared_disk_nothing
Share Nothing: Advantages
• Nodes that may go down or become unavailable to the network do
not slow down or inhibit any other node in the system.
• Subsequently, no true bottlenecks occur since no node has to “wait”
for another.
• Since nodes are completely independent right from the start, more
can be added to the system without the need to configure any other
existing node.
• This leads to systems with very good scalability. System can grow as
needed without significant interruption.
Publisher - Subscriber
• Reduce complexity between producers and consumers
• Solves problem of consumers receiving irrelevant
information/messages
• Solution is to put information into topics/subjects and allow
a consumer to subscribe to desired subjects
- Reduce code
- Reduces communication interactions
• Actors :
– Publisher
– Subscriber
– Communication Infrastructure (optional)
http://msdn.microsoft.com/en-us/library/ff649664.aspx
Rule Evaluation Architecture
CS 4311
Rule Evaluation Architecture
• Inference engine parses user input and
determines whether it is a fact/rule or a
query. If it is a fact/rule, it adds this
entry to the knowledge base.
Otherwise, it queries the knowledge
base for applicable rules and attempts
to resolve the query/goal.
• Also known as rule-based/expert
system architecture.
• It is a type of layered architecture
Rule Evaluation Architecture
• Solve problems understood as sets of predicates
– Iteratively explores problems whose general
solution approach is unclear
• Modifiability/Maintainability
– Behavior of the application can easily be modified
through dynamic addition or deletion of rules from
the knowledge base
• Reusability
– By only changing the facts/rules the system can be
used to solve problems in various domains
• Flexibility
– Additional facts/rules are easy to add
Rule Evaluation Architecture
• Actors
– User interface
– Inference engine
– Knowledge base (shared memory)
Rule Evaluation Architecture
• How do they relate to each other?
– Components are tightly interconnected, with direct
procedure calls and/or shared data access
– User interface provides two modes, one for adding rules
to knowledge base and another to query knowledge
base
– Inference engine operates on knowledge base in
response to user input.
– Facts and production rules are added to the knowledge base; goals
are compared against existing facts in the database. If an exact
match is found, it returns true to the user interface. Otherwise, it
evaluates the rules as necessary to determine the validity of the
goal.
Rule Evaluation Architecture
• Drawbacks
– When a large number of rules are involved,
understanding the interactions between multiple rules
affected by the same facts can become very difficult.
Understanding the logical basis for a computed result
can be as important as the result itself.
Rule Evaluation Architecture
Example: Lunar Lander in a rule-based style
To see what the status of the spacecraft is, the
user can switch to the goal mode and asks
whether the spacecraft has landed safely
:landed(spacecraft)
To handle this query, the inference engine
queries the database. If the existing facts and
other production rules satisfy the conditions
"altitude < = 0" and "velocity < 3 ft/s", then
the engine returns true to the user interface.
Otherwise, it returns false.
Rule Evaluation Architecture
Source: Taylor, R. et al., Software Architecture: Foundations, Theory, and
Practice. New Jersey: John Wiley & Sons, Inc., 2010