| Title: | DIGITAL UNIX (FORMERLY KNOWN AS DEC OSF/1) |
| Notice: | Welcome to the Digital UNIX Conference |
| Moderator: | SMURF::DENHAM |
| Created: | Thu Mar 16 1995 |
| Last Modified: | Fri Jun 06 1997 |
| Last Successful Update: | Fri Jun 06 1997 |
| Number of topics: | 10068 |
| Total number of notes: | 35879 |
During a presentation on AlphaServers I was asked to provide information regarding Digital's suggested resources for system management and operations for a typical 4100 server. Before you throw those darts you have just reached for, I agree that this is an open ended and need-to-qualify question. However, do any of our competitors (or us, for that matter) put out any guidelines (rough though they may be) that estimate what it will take for a particular server to be managed based on the O/S? In other words, how much % of a system manager will it take to manage a Digital UNIX 4.0b system? For planning purposes, this could be important (e.g. "we will need 2.5 FTE for managing the AlphaSERVERS running Digital UNIX 4.0+). This kind of makes sense because we should be able to quantify the ease-of-use we tout especially with the new system management tools in Digital UNIX. While much of this is OS specific, some is not - an AlphaServer is an AlphaServer. Any Comments? Regards, -BAP
| T.R | Title | User | Personal Name | Date | Lines |
|---|---|---|---|---|---|
| 8777.1 | KITCHE::schott | Eric R. Schott USG Product Management | Mon Feb 10 1997 20:21 | 33 | |
Hi
It depends a lot on :
- what application?
- is the workload steady state or irregular?
- Do the admin's configuring the machine understand how to automate
their common tasks to avoid day-to-day playing?
- How is the system configured...is all storage raid protected?
do they use LSM to mirror the disks?
- Are they running decevent regularly...
- how much do they have to change on a day to day to meet operational
needs (like add/remove users, or adjust storage usage).
In general, the base system once configured should run for quite
a while with little or no feeding assuming no hardware breaks...when
hardware breaks (like a disk), this can be a non issue (if using
LSM) or a major system mgmt headache (if you have to re-install).
My guess is it could be as little as .1 engineers (if they are
real good), to xxxx engineers
(how many xxxx does it take to change a light bulb...)
If they hire services to run the system, it would be their time to
manage services....
Eric
| |||||