Reference: Sommerville, Software Engineering, 10 ed., Chapter 17
High Level View
Virtually all large computer-based systems are now distributed systems. A distributed system is "a collection of independent computers that appears to the user as a single coherent system." Information processing is distributed over several computers rather than confined to a single machine. Distributed software engineering is therefore very important for enterprise computing systems.
Benefits of distributed systems:
Distributed systems and Design Issues
Distributed systems are more complex than systems that run on a single processor. Complexity arises because different parts of the system are independently managed as is the network. There is no single authority in charge of the system so top-down control is impossible.
Ideally, users should not be aware that a system is distributed, and services should be independent of distribution characteristics. In practice, this is impossible because parts of the system are independently managed and because of network delays. Often better to make users aware of distribution so that they can cope with problems. To achieve transparency, resources should be abstracted and addressed logically rather than physically. Middleware maps logical to physical resources.
Open distributed systems are systems that are built according to generally accepted standards. Components from any supplier can be integrated into the system and can inter-operate with the other system components. Openness implies that system components can be independently developed in any programming language and, if these conform to standards, they will work with other components. Web service standards for service-oriented architectures were developed to be open standards.
The scalability of a system reflects its ability to deliver a high-quality service as demands on the system increase:
There is a distinction between scaling-up and scaling-out. Scaling up is more powerful system; scaling out is more system instances.
When a system is distributed, the number of ways that the system may be attacked is significantly increased, compared to centralized systems. If a part of the system is successfully attacked then the attacker may be able to use this as a 'back door' into other parts of the system. Difficulties in a distributed system arise because different organizations may own parts of the system. These organizations may have mutually incompatible security policies and security mechanisms. The types of attack that a distributed system must defend itself against are:
Quality of service
The quality of service (QoS) offered by a distributed system reflects the system's ability to deliver its services dependably and with a response time and throughput that is acceptable to its users. Quality of service is particularly critical when the system is dealing with time-critical data such as sound or video streams. In these circumstances, if the quality of service falls below a threshold value then the sound or video may become so degraded that it is impossible to understand.
In a distributed system, it is inevitable that failures will occur, so the system has to be designed to be resilient to these failures. "You know that you have a distributed system when the crash of a system that you've never heard of stops you getting any work done." Distributed systems should include mechanisms for discovering if a component of the system has failed, should continue to deliver as many services as possible in spite of that failure and, as far as possible, automatically recover from the failure.
Two types of interaction between components in a distributed system.
1-Procedural interaction, where one computer calls on a known service offered by another computer and waits for a response.
Procedural communication in a distributed system is implemented using remote procedure calls (RPC). In a remote procedure call, one component calls another component as if it was a local procedure or method. The middleware in the system intercepts this call and passes it to a remote component. This carries out the required computation and, via the middleware, returns the result to the calling component. A problem with RPCs is that the caller and the callee need to be available at the time of the communication, and they must know how to refer to each other.
2-Message-based interaction involves the sending computer sending information about what is required to another computer. There is no necessity to wait for a response.
Message-based interaction normally involves one component creating a message that details the services required from another component. Through the system middleware, this is sent to the receiving component. The receiver parses the message, carries out the computations and creates a message for the sending component with the required results. In a message-based approach, it is not necessary for the sender and receiver of the message to be aware of each other. They simple communicate with the middleware.
Middleware is software that can manage diverse components of a distributed system and ensure that they can communicate and exchange data. The components in a distributed system may be implemented in different programming languages and may execute on completely different types of processor. Models of data, information representation and protocols for communication may all be different.
Interaction support, where the middleware coordinates interactions between different components in the system. The middleware provides location transparency in that it isn't necessary for components to know the physical locations of other components. The provision of common services, where the middleware provides reusable implementations of services that may be required by several components in the distributed system. By using these common services, components can easily inter-operate and provide user services in a consistent way.