On Tue, Apr 3, 2012 at 8:36 AM, Cat Okita <[email protected]> wrote: > On Tue, 3 Apr 2012, ?leen Frisch wrote: >> >> Jahidur Rahman, one of the students in the Network and System >> Administration Masters degree program needs input from experienced system >> administrators as part of his thesis research. Please consider taking a very >> few minutes to respond to his brief survey (link below). Thanks from him and >> me in advance. > > While I'm certain that it's a typo, I love the idea of a "Configuration > Management Toll" (and suspect that anybody that uses them feels the > same way ;>). > > That said, I'm a tad puzzled by having to select a single configuration > management product, especially gven the range of products presented, > which cover both servers and networks.
It's tricky to do these sorts of surveys, and it's usually the case that a user in the field is going to have RADICALLY different takes on 'appropriate' questions and their answers than a junior person (or a student, or someone who's not totally on fire about CM...) I immediately want to know what the student is trying to measure, and what their statistical goals are. I'm probably a bit weird in that sense. Several things about the survey kind of bug me - like that some of the 5-option answers look more like a 3-point Likert scale with 2 other options bolted onto one end. It's not so much a continuum.... I figure that the supervising faculty will ream the student (or not...) on their statistical prowess, at their whim. The analysis of the results from the survey should be fun... if the student is a bit sadistic. ;) > IME, it's pretty rare to only have one configuration management tool > in house -- and even less common to use the same tool (rather than > related tools or different tools) for both systems and networks, when > dealing with even modest degrees of scale. Bingo. We have similar issues when people ask us about our monitoring tools - I can think of five or six different things that we use, actively, that one could interpret as "monitoring tools" - it's just not a simple area to scope out. (And, well, neither is CM - ...) > Beyond that, I'd strongly suggest that it's important to include some > sort of information about the number of servers and or networking devices > being managed -- scalability is related, but not necessarily the same > thing (I can cheerfully say that a given product scales into the hundreds > of thousands, while at a site that has a grand total of 20 servers and > 2 networking devices). Number, type, number of elements on each one being managed - one wants to know whether the real element of scalability here is the 'server', or whether it's actually something extrapolated from the number of nodes and the number of elements on each node that you're managing..... --e _______________________________________________ Discuss mailing list [email protected] https://lists.lopsa.org/cgi-bin/mailman/listinfo/discuss This list provided by the League of Professional System Administrators http://lopsa.org/
