Hi,
I agree with you totally, but Fred is looking for a software only solution, as he refuses to talk abt hardware, which indicates he spec'ed & supplied the hardware & he is wasting a lot of peoples time looking for a configuration/software solution.
I have been called in to solve 8 similar problems in the last 12 Months, 4 of which we had quoted on, and were told (quite abusively in some cases) we were too expensive, you are trying to rip us off, only to be called in 6 Months later to fix the problems, as, in this case the hardware was totally inadequate.
The extra cost of the new hardware is not what the client wants to hear as they were originally told they could do it on an entry level server, a difficult situation to overcome. The minimum hardware spec's I gave him were what I used to overcome the problems in the above cases, as these clients were using the server for other processes as well.

Regards N Mc Donald

P.S. We made a lot more profit out of fixing the problem than the original supply quote, maybe this is a better way to go.

>Hi Neil,
>
>I agree with you that the hardware is under spec for this project. What is not stated in the problem is management’s willingness to spend money to solve the problem. Fred and his team cannot determine the size of the problem until current performance is measured. Then using management based on facts and data they determine the best solution possible.
>What would be the specs for the new server?
>Questions, which need to be looked at, are:
>1. What is the current workload?
>2. What services are going to run on this server?
>3. How much has this workload increased over a specific time?
>4. Are there any additional factors to this server’s performance?
>A classic example, we have a data server, which is running multiple database application, and running print services for the whole company. In the beginning the print services was only a few printers. However, over the year and half since the decision implemented, the number of printer has grown by 40 times. Yet management had the perception that because we purchased a “kick _ _ _” server there is no problem. However upon reviewing, the data collected during the base lining of the server’s performance we where able to change management perception of the server. Then using the data we where able to justify the monetary expenditure for the more memory,hard drives,etc,. Then we where able to justify the building of another server to handle the print server responsibilities. Sometimes when requesting money to solve problems you have to prove the problem exist and have the data to back up the request.
>
>
>Andy