December, 2008

RFG clarifies its clarification

Analyst firm Robert Frances Group is doing a lot of clarifications.

Last month Hewlett-Packard announced that 250 mainframe users had migrated to Itanium in the last two years, and cited an RFG study in the release. A quote from the HP release referencing the RFG study:

According to Robert Frances Group, a leading provider of consulting and research, the capital savings range from $1.5 million up to $23 million, with up to an additional operational cost savings of more than $4 million over four years. The study shows HP Integrity servers consumed 41 percent less energy and used 48 percent less space than the IBM z9 mainframes they replaced.

RFG piped up, saying readers could infer that Itanium is always better than mainframe, which is certainly not what RFG was trying to say. We wrote about it and had a headline that said RFG was “dissing” the HP announcement. A quote from the first RFG statement:

Last week HP issued a press release that referenced this report in which readers can infer that RFG believes a new HP system is always less expensive than a new comparable IBM System z. The report did not make such an analysis and made no comparative statements of that nature.

This is a misleading conclusion that RFG does not support. RFG has long stated that the mainframe is one of the best and most energy efficient platform options. RFG has written many research reports stating that mainframes should be considered and used in certain environments and RFG stands by those statements.

Well, now RFG has spoken again. In another clarification, the analyst firm says that the HP announcement never overstated RFG’s published findings. Here’s the full statement:

In the RFG press release on Nov. 17 “RFG Clarifies Its Position on System z Mainframe Pricing and Competitiveness” RFG attempted to clarify that the case study report referenced in the HP Nov. 10th press release compared current Integrity systems with a previous model, IBM z9 mainframe, not a new, IBM z10 mainframe. Further, RFG acknowledges that the RFG study information used in HP’s press release was accurate and factual and did not overstate RFG’s published findings. RFG stands by its findings. RFG’s sole concern was that readers would infer from the HP press release entitled “Business Customers Find Mainframe Costs Out of Step with Data Center Budgets” that RFG believes mainframes are non-competitive in all circumstances.

So it appears the issue has come down to the difference between “imply” and “infer.” Did HP imply that the RFG study claimed Itanium always trumped the mainframe, or was it just the wording of the release that could lead readers to infer such? And what is the difference?

Debugging Tip: The error is probably in your code

One of the answers to Robert Crawford’s first Assembler quiz included this story from Bruce Roy at the University of Connecticut:

I can’t remember who complained but I remember that he was kind of sharp guy doing some research on our system. He actually said that there was a bug in the Load Address instruction. Peter Furmo, who received the complaint, took it seriously for about 10 seconds. Then, he reasoned that if there was a bug in that instruction, OS/MVT would not be running which, of course, was not the case.

Crawford’s moral of this story: The error is probably in your code.

Wordles: System z10, Superdome, Sun M9000

At a site called Wordle, you can input a text document and out comes what Wordle calls a “word cloud.” It is basically a visual representation of the text document, with the most often used words represented larger and the least often used words represented smaller.

So I decided that for kicks, I would create a Wordle from the System z10 Enterprise Class reference guide. I downloaded it, converted it to a text document, and pasted the text into Wordle. This is what it spit out (click the image for the slightly larger version):

So then I thought it would be a good idea if I did a similar thing with the mainframe’s competitors. So I picked out the HP Superdome and the Sun Microsystems M9000 and did the same thing. Here’s the Superdome:

And the Sun M9000:

It was interesting to see what each company stressed in their enterprise server documentation, other than the name of their own product. For example:

IBM z10: Support, FICON, data, capacity

HP Superdome:  Chassis, system, partition, cabinet

Sun M9000: Power, CPU, Enterprise, memory

Crawford’s Mainframe Assembler Quiz, answers and part 2

In his October mainframe column, columnist Robert Crawford asked what commonly used Assembler instruction can be used to add three numbers together and place the answer in a fourth location. The answer is Load Address (LA).

Below is Crawford’s answer and the next Assembler Quiz.

Most of us have seen LA used to get the address of a label:

LA R13,SAVEAREA

But if you look in the Principle of Operations manual you will see the actual format of the instruction is as follows:

LA R1,O(X,B)

Where: R1 is the register to receive the address
X is the index register
B is the base register
O is an offset to be added to the sum of the base and index registers

In other words, the magical thing we call an address is actually just a number computed by adding two registers together with an offset. So, at the end of this instruction sequence:
LA R1,4 R1=4
LA R2,3 R2=3
LA R3,2(R1,R2) R3=2+4+3

Register 3 contains 9

We received 9 correct answers to this question. The first three will receive a Startbucks gift card.

New Assembler Quiz: One more quiz before I run out of ideas. What sequence of three instructions can exchange the values of two registers without using memory or a third register? As before, send your answers to editor@searchdatacenter.com and the first three correct answers will win a $5 Starbucks gift card.