FUD or not FUD?
I think IBM marketing is best. According to wikipedia article on "FUD", IBM was the first company to embrace FUD in a wide scale. Here we have some more IBM examples:
I really like when IBM claim they can consolidate many computers onto a single IBM machine. It is like their Mainframes: IBM can apparently consolidate 1500 x86 server to a Mainframe. What IBM does not tell you (you have to read carefully) is that each x86 server is utilized at a few percent and the Mainframe has close to 100% utilization. With the same weird reasoning, I can utilize a couple of idling x86 servers on my iPhone. Ergo "iPhone can virtualize several x86 servers!!!". True statement or untrue? FUD?
Also, IBM does not tell you that IBM Mainframe cpus are dog slow. 16 Nehalem-EX cpus can easily match one z10 Mainframe with 64 cpus when we talk about raw cpu performance. So how can 1500 x86 be utilized on some 10ish crappy Mainframe cpus? What happens if the x86 servers start to do some cpu intensive work?
I would not be surprised if the x86 servers that IBM wants to consolidated onto a Mainframe, are 1GHz Pentium 3 or even older. IBM loves to compare new h/w to old - giving the impression of superiority.
.
Regarding this POWER7 consolidation, it is the same. Apparently IBM can consolidate many machines onto a POWER7 box. But IBM does the same old trick again
1) Compare to very old machines with bad performance
2) Assume the old machines have low utilization whereas the IBM box has high utilization.
I wonder how IBM would react to this advertisement from Intel?
"Now we can consolidate 125 POWER boxes onto a 8 socket Nehalem-EX machine! Save money and power! x86 is much faster than POWER" - and Intel dont tell anyone that the POWER boxes are all old idling POWER5, and the Nehalem-EX box is fully utilized. Do you think IBMers would cry out loud and call this FUD? Or would they accept this as a fact, gladly? Would this statement be FUD or not?
.
Earlier I complained on a Sun comparison of an Sun box to an inferior HP box. I asked Sun to withdraw the benchmark. I wrote
"How can you compare HP 4 dual core 1.6GHz vs SUN 4 quad 2.53GHz? Even a dog would understand that it is not fair. This is something IBM could do, but SUN didnt. I thought? But maybe I am wrong?"
http://blogs.sun.com/mandalika/entry/peoplesoft_na_payroll_240k_ee
And in the article on this site here, I am proven correct. To compare new h/w to old h/w is typical for IBM, to try to give the impression of superiority. Not fair play in my point of view. But maybe IBM think this is fair? IBMers would never try to fight a guy in their same size, they only fight smaller guys, and only if the smaller guy is tied up. Fair play does not exist.