Yikes! You must have big plums to persist with a customer like that. It sounds like a disaster waiting to happen!
Symon -----Original Message----- From: The Long and Winding Road [mailto:[EMAIL PROTECTED] Sent: 08 March 2003 19:44 To: [EMAIL PROTECTED] Subject: Network Design - What Priscilla did NOT cover in her book: WAS [7:64842] ""Symon Thurlow"" wrote in message news:[EMAIL PROTECTED] > Hey Chuck, > > How did that big design go, the one you mentioned on the list a few > months ago? > > Symon You mean the Never Ending Design? The Nightmare before the CCIE Lab? Here is a brief rundown. I will say in advance that as all of you who work in the real world with real world management, real world customers, and real world situations already know, the real work is at layers 8,9, and 10. Project Summary: large organization, 2000+ employees, 10,000 data ports, 3 dozen locations, with each location being a campus of several buildings or several floors within buildings. The project RFP called for a complete forklift of the existing infrastructure - routers, switches, PBX. It also called for wireless for voice and data. The project goal was to create a network fully capable of providing seamless integrated services for data, voice, and video. Oh yes, there was a three week turnaround deadline for the response, and there was no flexibility in this. Meet the customer date or lose the opportunity. On top of that, as is typical with most RFP's, all questions are to be submitted in writing, and all responses go to all bidders. Clues that something is strange: 1) for any wireless response this complex, detailed site surveys are required. there is not time to do this. answer: well then just do a site survey. besides, we have aerial photographs of all of our locations posted on our web site. you can use those to determine what you need. 2) you're RFP provides numbers of IDF's in each location and total number of ports required. e.g. site X has 7 IDF's and 257 data ports. do you have detail as to how many data ports are in each specific closet? answer: use an average, or come out here and do a site survey and figure it out for yourself. 3) you're RFP calls for L3 switching in each and every closet. Is this necessary, given that there is only a single ingress/egress, and that all sites are hub and spoke? plus L3 is more expensive, and I'm not sure there is anything to gain. answer: we want L3 everywhere. are you saying your ( Cisco ) equipment does not do L3? Customer: oh by the way, we will be opening a new location sometime in the next 18 months. I want you to include that location in this response. 4) how many closets? how many phones? how many data ports? answer: just take locations a,b, and c, and average those out to get the numbers. These were the major things, and should give you a pretty good idea of the upper layer issues. Well, I work my ass off to meet the deadlines. We and a couple of other vendors respond. The presentation meeting takes place with all vendors in the same room at the same time. Oh joy, but at least we can see eachothers' hands. All vendors come back with total cost in the 8-9 million range. Now the customer reveals that his budget is 5 million. This is something that was asked, and which the customer refused to discuss previously. I should add that as this is a non profit organization, and some of the funding is from grant money with particular restrictions, this is not as straightforward in terms of budget as might first appear. The grants will pay for some types of equipment and services, but not others. The 5 mil is for a "complete package" including data circuits, all equipment, and all services. so subtract the total 5 year cost of data circuits from that 5 mil. divvy up what's left between what the grants will buy and what the customer himself will buy. OK, so now we have to scramble. The customer finally gets a clue that things cost money, and the more you want, the more you have to pay. So - trim your proposals, and get back with just what is required for end to end voice over IP plus new WAN equipment. No wireless. No new switches other than those needed to directly support the IP telephones. back to the drawing board. All non-phone switches are out. all wireless is out. next big problem. the customer RFP states specifically that there are numbers of site with poor wiring, and inadequate equipment. There are express concerns with the ability of existing infrastructures to handle existing loads, let alone adding unified messaging to the mix. we suggest using a voicemail only solution. the customer goes into apoplexy. my network is my business, not yours. well, what if performance suffers and you end up with unhappy and complaining users. well that's my responsibility and none of your business. OK. we all know what's gonna happen, but ok. In the mean time, one of my fellow workers is doing physical site surveys. Among the things he discovers is an additional 21 data closets that the customer was unaware of. the numbers of data closets as expressed in the RFP is wrong. Many sites have one or two fewer. Other sites have as many as 6 or 7 more. scramble again to change the design to reflect this.all the time under this damn budget restriction. The customer will not hear of doing this over a couple of years, obtaining addition grant many in future funding periods. the customer will not hear of further reductions. The customer is pissed that I have to resort to single routers in many locations, routers which will serve as PSTN failover, PSTN gateways, and WAN routing all in one box. The customer says that Cisco told them that AVVID is a redundant solution and he wants redundancy. I reply that I have to be concerned with two things - ability of the router to handle the peak load demands for voice and data, and the budget, which is a bummer. I say I can go cheaper, but then I risk having to up the router later if it proves inadequate. Response: I'm buying a managed solution, and if equipment proves inadequate, you will replace it at no cost to me. Does this customer scare you? He sure scares me. Notice that all of the discussion from the customer side has little to do with needs and requirements, and everything to do with wants and demands. Notice too the responsibility issue. In any case, as it stands today, a couple of design revision later, we have something that will work. I am not comfortable with the lack of failover for the Unity box, supporting over 2,000 voice mailboxes. I am not comfortable with the LAN issue, because I am still responsible for the WAN and I have no control over the customer LAN. Worse of all, I have no confidence that this customer really will accept responsibility for the things he said were none of my business. I see a major disaster coming down the road. Oh - I see I haven't even mentioned the phones, the phone requirements, and what was eventually the compromise there. On the other hand, the commission for this sale will be decent :-> ============================================= This email has been content filtered and subject to spam filtering. If you consider this email is unsolicited please forward the email to [EMAIL PROTECTED] and request that the sender's domain be blocked from sending any further emails. ============================================= Message Posted at: http://www.groupstudy.com/form/read.php?f=7&i=64851&t=64851 -------------------------------------------------- FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
