[cones688]

I think you are underestimating the scale of what a Z system can run... You might not have a use but every fortune 500 and banking institution in the world does - and they need it to work, 5 9's isn't good enough.

> IBM Hursley laboratory director Rob Lamb says: “There are 6,900 tweets, 30,000 Facebook likes and 60,000 Google searches per second." The mainframe CICS runs 1.1m transactions per second, which equates to 10bn per day [0]

[0] http://www.computerweekly.com/feature/Can-the-mainframe-rema...

[hueving]

That equivalence is bogus. If a search was as simple as a single CICS transaction, Google would just run that and be done.

Mainframes are overpriced and inefficient, but they are the only option for a F500 without the in-house talent to build any kind of distributed, fault tolerant system.

[jussij]

If as you say mainframes are overpriced and inefficient, that means there is a great opportunity for some organization to move with a much cheaper, more efficient option.

But people have been predicting the death of big iron for several decades now, yet they live on.

I don't doubt mainframes might be overpriced, but I also suspect the reason they persist is they have yet to come up with a cheaper option, offering the same performance figures.

[graycat]

Long one of the main reasons for IBM mainframes was the bet your business software that wouldn't run anywhere else and that would be too expensive to rewrite to run somewhere else.

Also, there is a remark that in major parts of the financial industry, running an IBM mainframe is nearly a necessary condition for compliance.

[jussij]

> would be too expensive to rewrite to run somewhere else.

I'm don't doubt that is a major factor. Add to that the major risk that what every new system you move to might actually fail to work or end up costing more.

[toomuchtodo]

> If as you say mainframes are overpriced and inefficient, that means there is a great opportunity for some organization to move with a much cheaper, more efficient option.

But isn't that what Facebook, Google, and Amazon are doing? Using massively distributed commodity x86 hardware to eat away at incumbent businesses that would outsource their IT services to mainframes? Last I read, Google is about to go into auto insurance, and all three companies I listed do payment processing.

If software is eating the world, SV behemoths are eating business verticals.

[sytelus]

The electrical power ($600/day vs $32/day), floor space (10,000 sq ft vs 400 sq ft), and cooling costs for those mainframes were less than those of distributed servers handling a comparable load. In addition, those mainframes required 80 percent less administration/labor (>25 people vs <5 people); “Mean Time Between Failure” measured in decades for mainframe vs months for other servers.

96 of the world’s top 100 banks, 23 of the 25 top US retailers, and 9 out of 10 of the world’s largest insurance companies run System z Seventy-one percent of global Fortune 500 companies are System z clients Nine out of the top 10 global life and health insurance providers process their high-volume transactions on a System z mainframe Mainframes process roughly 30 billion business transactions per day, including most major credit card transactions and stock trades, money transfers, manufacturing processes, and ERP systems.

The new mainframe delivered in 2010 improved single system image performance by 60 percent, while keeping within the same energy envelope when compared to previous generations. And the newest mainframe which shipped in 2012 has up to 50 percent more total system capacity, as well as availability and security enhancements.

It uses 5.5 GHz hexa-core chips – hardly old technology. It is scalable to 120 cores with 3 terabytes of memory.

[1]

[1] http://www.share.org/p/bl/et/blogid=2&blogaid=234

[mimighost]

A single search on Google could lead to 10x-100x actual requests to the backend system. So the comparison here is not really a fair game.

Second, since I assume, most of the programmers here don't have access to mainframes, it is hard to testify those numbers. Also, since it is a benchmark it will be useful to revel what exactly the task they are using here, otherwise I would simply throw this claim into my 'pure PR mess, don't take it seriously' bin :)

[graycat]

CICS itself is essentially just a way to put up the forms for a user interface, but a CICS application usually makes heavy use of database, say, relational database, e.g., DB/2 although there may still be some IMS usage still hanging on.

One use of CICS was for heads down medical claims processing, across all four US time zones. The site our team from IBM Research visited wanted high reliability: If the site was down for, say, an hour, then the data entry staff would have to be called back on a Saturday, for at least half a day, at a higher rate per hour. One such outage in a year, and the CIO could lose his bonus. Two and he could lose his job. The site was very uptight. Getting into the glass house was not easy; might have been easier to get into the White House Oval Office.

At one time to make CICS more secure, there was some interest in having processor hardware support for address sub-spaces. Another idea was cross memory where a program could call and execute, say, a function in another address space. There were also data spaces, that is, address spaces with just data and no code but that could be accessed by other address spaces with code.

Net, the IBM mainframes are not really simple things. Cloning one would not be easy, and at IBM's next version of hard/software, the clone could be unable to run the newer software and suddenly be a boat anchor.

[bane]

I have practically zero experience with mainframes, but I've always heard about their insane throughput figures. I've never really seen much on their architecture. Any insight in how they work and do so much?

[filereaper]

Here's an old 2012 article about zEC12 (previous) generation of the mainframe. http://www.theregister.co.uk/2012/09/05/ibm_z12_mainframe_en... zEC13 was just released this year.

The mainframes have their own CISC type architecture and typically have massive amounts of cache and a high clock speed (5Ghz and above)

Their instructions sets are also different in that its not strictly of the von-Neumann variety. Mainframes can do things like memory copies directly in memory without requiring copying via registers. This kind-of attacks the von-Neumann bottleneck directly and is good for batch and high volume transaction processing.

The software for mainframes is also typically fused into their kernels, things like CICS and DB2 are not "user" programs, they're part of the OS so I/O is handled much better and things don't "block" as much.