In publish-subscribe systems, explain how channel-based approaches can trivially be implemented
using a group communication service? Why is this a less optimal strategy for implementing a
content-based approach?

What will be an ideal response?

In a channel-based approach, events are published to a named channel and subscribers subscribe to a given
channel and receive all messages sent on that channel. This can be implemented directly using group
communication, that is when a channel is created, you create an associated group; subscriptions are
implemented through joining the group and then subsequent events from publishers are sent to the group and
hence delivered to all subscribers who are then members of the group.
This trivial mapping is not possible with content-based approaches where communication is more
associative and based on content (more specifically, based on arbitrary queries expressed over the content).
This cannot easily be implemented using group communication. One option would be to send all events over
one group and then do the required filtering in each subscriber, but this would be really inefficient. In practice,
as we have seen in Section 6.3.2, efficient implementation of content-based approaches requires the creation
of an suitable overlay implementing a content-based routing algorithm, that is routing is influenced by content.

Computer Science & Information Technology

You might also like to view...

Compare the three schema designs in four categories: global and local declarations; nesting of elements; reusability; and interaction with namespaces.

What will be an ideal response?

Computer Science & Information Technology

Answer the following statements true (T) or false (F)

1. Data structures are those data items within the system that are not meaningful if broken down further. 2. A default value is one that never changes on a report. 3. A discrete data element is one that has certain fixed values. 4. When a V is used as a formatting character for an element, it indicates where the decimal point should occur, even though the actual decimal point is not included. 5. All derived elements must be stored on a data store.

Computer Science & Information Technology