11.6 Negotiating Support

< Day Day Up >

In the complex corporate environment, most support is rendered through centralized operations that have specific capabilities or lack thereof. The support profile can vary from data center to data center should you work in a big organization. Differences can be found in technologies supported, service levels, and ease of physical access for project engineers. Regarding this last point, keep in mind that the operations team is rarely looking for new customers. So your arriving on their doorstep at the last minute with no paperwork, and worse, no prior consultation or schmoozing with the appropriate individuals, is a good way to delay your production turnups.

Just as project managers are often seen as free-spirited and arrogant, data center supervisors can be pretty inflexible unless you are willing to play by their rules. Not that this is always possible, but it is worth making the effort. I have found the best way to handle this is to reach out and identify the key individuals, well in advance of your critical dates. Introduce yourself, describe your project briefly, tell them what you are looking to do, and ask how that gets done. I have had varying degrees of success with the following data center support management types:

Implementation manager. Most facilities have an intake team to transition equipment into the environment, similar to an internal project manager. They are possibly technology specific (i.e., mainframe, midrange, etc.).
Network/telecom. This team is responsible for the management of the LAN and WAN devices and connections, as well as the assignment and configuration of Internet Protocol (IP) addresses and switch ports.
Voice/telecom. Some data centers support voice applications that use interactive voice response (IVR) or computer telephony integration (CTI) technologies. These applications integrate voice and computer functions.
Hardware implementation. This team manages the real estate in which computing platforms are installed (i.e., floor space, racks, and cabinets).
Hardware repair. This is the break-fix team that is also responsible for hardware upgrades such as adding storage, dedicated tape units, or network interface cards.
OS. Baseline support is provided for the various operating systems you will find in a full service data center. This team probably also handles things such as security or antivirus software.
Monitoring. Monitoring tools that trap Simple Network Management Protocol (SNMP)-type alerts regarding basic hardware health are increasingly common in all computing platforms these days.
Web masters. A lot of shops have teams dedicated to supporting the various "canned" Web server systems and tools.
Database administrators (DBAs). DBAs are commonly deployed to support these repositories of corporate intellectual property and operating data for the name brand database management software.
Change control. Most data centers allow changes to applications or systems during specific maintenence windows, and may in fact be the only personnel allowed to initiate upgrades to application executables or database table structures. Details on this vary. Normally, some sort of online process is in place whereby you request changes weeks in advance of your targeted date.
Storage management. Long a mainframe specialty, this technology has proliferated with devices such as storage area networks (SANs) and network attached storage (NAS) into the distributed processing world. This team sometimes has disaster recovery responsibilities as well.
Tape and archive. This team manages the backing up of data and systems as well as the physical management of media for offsite storage, rotations, and transport to disaster recovery (DR) processing centers.
Security/demilitarized zone (DMZ). Some applications support users outside the corporate network, be they a mobile workforce, customers, or trading partners. If your project is in this area, you will undoubtedly have concerns regarding firewalls and proxy or reverse proxy servers, as well as authentication, encryption, and other related technologies.
Disaster recovery. This team is responsible for coordination and testing of DR requirements of business critical applications.
Call center. Most troubles, whether detected via monitoring or user-reported, are managed through a call center that opens trouble tickets, assigns resolvers, and follows up with support, development, or user personnel as required or as dictated by corporate IT policy.

When I compiled this list, I had no idea that the final team count would be as high as 15. Depending on the nature and complexity of the project, you may have to deal with many, if not all, of them. Keep in mind that these people are process-oriented. Precision and completeness in documentation is heavily enforced, and lead times are long, even after intense escalations. They may not appreciate your misapprehending the word "No." Sometimes, it appears that project managers do not understand the word "No," even though that appears to be support personnel's favorite word.

That statement may not seem fair to either side, but it points you in the right direction when reacting to the processes that will be thrown at you when you attempt to invade their inner sanctums. Hopefully, a few well-placed phone calls will smooth the way. You are always better off establishing relationships with the appropriate players in these teams as required. With any luck at all you will be working with them in the future, so you want to make sure you get and remain on a friendly basis with them. Chances are they have more leverage than you because it is unlikely you can wander into a data center with a pair of pliers and a couple of CDs and make anything you wish happen. This is not the old LAN room model.

< Day Day Up >