This white paper discusses the approach used to convert the Hotmail web server farm from UNIX to Windows 2000, and the reasons the features and techniques were chosen. It will focus primarily on the planners, developers, and system administrators. The purpose of the paper is to provide insight for similar deployments using Windows 2000. We will discuss the techniques from the viewpoint of human engineering as well as software engineering.
It would be interesting to read some similarly detailed docs that detail migrations from Windows to other OSes.
It doesn’t mention that the migration took more than three years: it didn’t take the Hotmail team that long to build the site in the first place.
What really struck about reading this — and I’m assuming it’s genuine — is how little the author understands about UNIX operations and administration. I don’t know any sysadmin who would use rsh to manage multiple systems, or even to login to another system.
No mention is made of cfengine or other methodologies that deal with managing clusters of machines of different classes.
And of course, there’s nothing to indicate that Windows is superior in design or implementation to the systems it’s replacing: in fact, there are repeated references to areas where FreeBSD et al are superior. The claims to better performance or better logging and monitoring strain credulity.
Hardware detection better in Windows? I doubt it, for anything that’s not a desktop/end user gadget.
As for it being “impossible” to detect a server thread loop that bound up a CPU on UNIX, I doubt that: while the tools available in UNIX might not be as well-packaged as Visual Studio, claims of their inadequacy stem from ignorance.
Sadly, some people will read this and be taken in by it.