Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                
Scribd
Upload
52 views

Distributed Systems Midsem Answers

A three-tier architecture is proposed for a car rental company's distributed system, with user interface, application, and data tiers. This improves performance by separating concerns, but adds complexity. In wide area networks, packets may arrive out of order due to variable speeds and reliability across networks. In local networks, packets are less likely to be reordered. When designing server/client communication, ports must use a standardized naming convention like IANA numbers for interoperability.

Uploaded by

MARTIN BOATENG
Copyright
© © All Rights Reserved
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
52 views

Distributed Systems Midsem Answers

A three-tier architecture is proposed for a car rental company's distributed system, with user interface, application, and data tiers. This improves performance by separating concerns, but adds complexity. In wide area networks, packets may arrive out of order due to variable speeds and reliability across networks. In local networks, packets are less likely to be reordered. When designing server/client communication, ports must use a standardized naming convention like IANA numbers for interoperability.

Uploaded by

MARTIN BOATENG
Copyright
© © All Rights Reserved
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 4

Distributed Systems Midsem Answers

1. Consider a hypothetical car hire company and sketch out a three-tier solution
to the provision of their underlying distributed car hire service. Use this to
illustrate the benefits and drawbacks of a three-tier solution considering
issues such as performance, scalability, dealing with failure and also
maintaining the software over?

Answer
The user interface tier would be responsible for providing the interface that
customers use to interact with the car hire service. This could include a
website or a mobile app that allows customers to search for and book cars, as
well as providing information about the cars and the company's policies.

The application tier would be responsible for managing the core business
logic of the car hire service. This could include handling requests from the
user interface tier, interfacing with the data tier to retrieve and store
information about cars and customers, and managing the allocation of cars to
customers.

The data tier would be responsible for storing and managing the data used
by the car hire service, such as information about cars, customers, and
bookings. This could be done using a database management system.

Benefit
One of the key benefits of a three-tier solution is that it can improve
performance by allowing different components of the system to be
optimized for their specific tasks

Drawback
One potential drawback of a three-tier solution is that it can be more
complex to design, implement, and maintain than a simpler system.

2. Explain how it is possible for a sequence of packets transmitted through a


wide area network to arrive at their destination in an order that differs from
that in which they were sent. Why can't this happen in a local network?
Answer
In a wide area network (WAN), packets may travel through multiple
intermediate networks and devices before reaching their destination. This
can lead to situations where packets are not transmitted in the same order in
which they were sent, due to differences in the speed or reliability of the
various intermediate networks and devices.

In contrast, in a local network, packets are typically transmitted over a


single, dedicated network connection, and are not subject to the same types
of delays and variations in reliability as in a WAN. As a result, packets are
much more likely to be transmitted and received in the same order in which
they were sent.

3. A server creates a port that it uses to receive requests from clients. Discuss
the design issues concerning the relationship between the name of this port
and the names used by clients.

Answer
One of the key design issues concerning the relationship between the name
of a server's port and the names used by clients is the need to ensure that the
server and clients can communicate using a shared, well-defined naming
convention. This is necessary in order for the server to know which port to
listen on and for clients to know which port to send requests to

One approach to addressing this issue is to use a well-known, standardized


naming convention for ports, such as the Internet Assigned Numbers
Authority (IANA) port numbers. This allows both the server and clients to
refer to ports using standardized names that are known to be unique and are
not likely to conflict with other port names.

4. Discuss whether message passing or DSM is preferable for fault-tolerant


applications.

Answer
Both message passing and distributed shared memory (DSM) are
technologies that can be used to implement fault-tolerant applications. The
choice between these two approaches depends on a number of factors,
including the specific requirements and constraints of the application, the
performance and scalability needs of the system, and the availability of
supporting infrastructure

DSM is preferable for applications that require a high degree of data sharing
and coordination, and that can tolerate a higher level of complexity and
overhead. This is because DSM provides a shared memory space that can be
accessed and updated by multiple components of the system, allowing for
more efficient communication and coordination. However, DSM typically
requires specialized hardware and software support, and can be more
complex to implement and maintain than message passing.

5. Assuming a DSM system is implemented in middleware without any


hardware support and in a platform-neutral manner. How would you deal
with the problem of differing data representations on heterogeneous
computers?

Answer
One approach to dealing with the problem of differing data representations
on heterogeneous computers in a DSM system implemented in middleware
without hardware support is to use platform-neutral data types and
serialization formats. This allows the DSM system to represent data in a
form that is independent of the underlying hardware and software platforms,
and can be easily transmitted and deserialized on different computers.

Another approach is to use data type conversion and translation techniques


to convert between different data representations on heterogeneous
computers. This can be done using mapping tables, conversion functions, or
other techniques that allow the DSM system to translate data from one
representation to another as needed.
6. If a communication paradigm is asynchronous, is it also time-uncoupled?
Explain your answer with an example as appropriate.

Answer
No, a communication paradigm being asynchronous does not necessarily
imply that it is also time-uncoupled.

Asynchronous communication refers to a mode of communication in which


the sender and receiver do not need to be actively connected at the same
time in order to exchange messages. This allows the sender and receiver to
operate independently, without the need for constant coordination or
synchronization.

Example: An example of an asynchronous but time-coupled communication


paradigm is email. In this case, the sender and receiver do not need to be
actively connected at the same time in order to exchange messages, but they
do need to operate at roughly the same speed in order for the communication
to be successful.

You might also like