|
Previous
|
Content
|
Next
|
|
| 3.0. Some final comments |
|
|
|
| To have a taste of QoS we have used some Cisco QoS tools and we are intending to stop here. Our main goal has been reached. Having some knowledge about QoS, what it means, why it need to be applied and how to do it. I hope this document could serve as starting point for you to continue investigating about this fascinating field of network management and its behavior optimization. |
| It's really a pity that most companies and technical people using Cisco routers just use them as a best-effort mechanism to forward traffic, without going deeper to have more from their investment. These routers have the capacity to do really a better job; it's just a matter that you invest a little of time and effort to learn how to configure them for getting better services for you, your users and customers. And, in many cases, you don't have to pay more because your routers have already installed the software code required to implement these services. This means that you could improve the quality of service you give without expending a lot of additional money. |
| Sometime companies invest unnecessarely more money to expand their infrastructure when perhaps, using QoS technology, this investment could be reduced or optimized by using better the current equipments they have. In short, it's just a matter to do more with less. |
| But not only Cisco has these tools available on its routers. Also Nortel, Alcatel and Juniper have them, that I know, at least. And Linux, being an open technology, has an amazing battery of really incredible tools already installed in its basic kernel. And with Linux the cost is almost nothing. Again, it's just a matter to investigate and learn how to use this offer. |
|
|
|
Don't be afraid of QoS. Anything is worst than the effort haven't done. Also is very important not
being so orthodox or so risky. Extremes are bad. Always test your ideas before putting them on production. Don't put on risk your network. Never, never, never use the current production network to implement new ideas and designs without having tested them before very well. In these cases you have to be orthodox. |
|
| But, when you are sure, really sure of what you are doing, don't be cowardly to implement your ideas. There are people that want to make things to improve their network behavior, but, they don't want to change anything. They argue: well, you know, I don't want to change this, also I don't want to replace this, and I wouldn't like to add this, and it's better not to touch this, and so, so, etc., etc. Think about it. If you don't do anything different that you do everyday then everything will be the same. Knowing exactly what you want, knowing exactly what you are doing, and knowing exactly up to where you can go is the better path to implement QoS. |
|
|
|
|
|
Previous
|
Content
|
Next
|