Not sure if this is of interest to anyone (I’d love to get feedback), but whenever I get asked to design a scalable hierarchy for a customer I run through these 20 or so questions with the consultant/customer to help me design a SMS2003 architecture based on a customer's needs/requirements. With this information I can then start to understand constraints on such items as time, environment (both physical and geopolitical) etc.

I thought I’d post these to get people thinking about the planning phase of a SMS2003 deployment.

SMS Feature Requirements:

·          What are expected SLA's for:

·          SINV, HINV, Patch Mgt scan, SWMetering reporting at central site from the time the client initiates the scan, from a reporting pint of view.

·          SWD SLAs and sizes of:

·          Patch/emergency SWD.

·          Regular SWD.

·          Large SWD (office, SP etc).

·          What is the expected SINV size and configuration of files to scan (will it be *.exe or a more focused scan (hint, hint)?

·          And at what frequency?

·          What is the expected HINV size and configuration (will they want to change the default def.mof)?

·          And at what frequency?

·          What is the expected SWMetering size and configuration (how many products/files etc)?

·          And at what frequency?

·          How many SWD advertisements will be created at the central site? In total and per week?

·          How many SWD packages will be created at the central site? In total and per week?

·          How many collections (and what size, dynamic or static) will be created at the central site? In total and per week?

·          How many advertisements per week/month per client will be initiated?

·          What’s the average number of clients expected to be targeted by a weekly or monthly SWD advertisement?

·          Will the customer be expected to report on SWD to client progress from the central site?

·          How many child sites (a basic typology would really help at this point). Include # of clients, WAN links etc.

·          What's the AD environment?

·          What discovery/client deployment methods they wish to use? And how often?

·          What's the backup strategy for this environment. What's the downtime SLA?

Standard Environment Details Needed:

·         Number of physical locations within the company.

·         Network links (both link speed and available speed).

·         Number of users/machines at each location.

·         Any geopolitical or legal limitations of design.

·         Define any reporting hierarchies. HQ reports data from 4 regional locations, who report for each of the locally based branches etc.

Basically a very detailed Visio document is worth a thousand words.