Although it can be rather difficult to network, the principle itself is quite simple. In order to communicate with each other, what a network does is connect devices such as machines, servers and printers together. For different kinds of cabling such as Ethernet or fibre optic cables or even wirelessly, this can be achieved.
With all the complicated switches, routers, firewalls and other equipment that we use to connect our devices in the workplace and even in other cities, states or even nations, things were much easier in the early days of networking than they are today. A peer to peer network has been considered one of the first kinds of networks to be configured. These are still in operation now although as you can find in most companies, they are not quite as popular as client-server networks. We will explain what each type of network is in this article and how they vary from each other.
With all the complicated switches, routers, firewalls and other equipment that we use to connect our devices in the workplace and even in other cities, states or even nations, things were much easier in the early days of networking than they are today. A peer to peer network has been considered one of the first kinds of networks to be configured. These are still in operation today, although as you’ll find in most organisations, they are not about as popular as client server networks. We will explain what each type of network is in this article and how they vary from each other.
Peer to Networks of Colleagues
The first form of network to be used was Peer to Peer Networks, also called workgroups. There is no unified administration or protection for this sort of network and each machine is responsible for its own local users and file and folder permissions. Any user who needs access to services on another computer would need to have an account on that particular computer and there is no unified user control.
So if Sally is a Computer A administrator and she wishes to access files on Laptop A and Computer C, then she would need to make a user account for an admin on Laptop A and Computer C and then grant the permissions that she requires to be able to access those tools. When the number of machines on the network rises, one can imagine how complex things will get.
When the number of computers in a peer-to-peer network begins to go above 10, issues such as slowdowns from network broadcasts and other traffic can be experienced because all the traffic goes to each computer, even though only the computer to which it was supposed to go can accept the data. Plus, certain operating systems installed by the workgroup will only support 10 concurrent connections at a time. So, if you have a file server machine and 20 people, then only 10 of them will connect at a time to the file server.
For home networks or small office networks where there are not a lot of users and machines to handle, peer to peer networks operate perfectly. But after you hit a certain cap, there is something else you need to introduce, such as a client-server network.
A client-server network has clients (workstations) as well as a server, just as the name suggests (or many servers). The clients are labelled Computer A, Computer B, Laptop A and so on in the diagram below. There is also a file server and a directory server used to manage access controls and user accounts.
Instead of a hub as we had in the peer to peer network, all computers and servers link to each other with a network switch, even if you can still use a switch for a peer to peer network. The key benefit here is that every user account on the directory server is generated and then each machine, laptop and other server is connected to a domain where login and resource permission authentication is centralised. A domain is a centralised way for machines, users and services to be handled and any computer enters the domain and each user is generated as a user of the domain. So if a user named Joe on computer C wishes to access Laptop B files, they can do so, assuming they are allowed to access their user account. In addition to the original user generated on the directory server, there is no need to make a user account for Joe on Laptop B or any other device on the network.
By having a switch instead of a hub, broadcast traffic is minimised because the switch knows the port to which each device is connecting and does not have to go to each computer or server to locate the one to which it is attempting to get. Switches can be thought of as “smart hubs or “dumb” switches can be thought of as hubs.
The client-server architecture is also very scalable and only the licence model and ultimately the hardware limitations in terms of network bandwidth and server availability are limited to the quantity of concurrent connections to a server. Directory servers can handle the administration of thousands of users with the few requisite hardware resources.
The client-server model’s downsides include:
- Increased costs because servers are more costly than hubs for machines and network switches.
- Because of the sophistication, it is more difficult to incorporate and manage.
- If a directory server goes offline and none of the users can log in, there is a single point of failure. This is also circumvented by getting several directory servers
Overall, you can go for the client-server model if your environment has multiple properties and you want to centralise control of your users and machines. If you just have a few machines and users, so it should work just fine with a peer to peer setup. Plus, if appropriate, you can reconfigure it into a client-server model in the future.