Slashdot Mirror

I count on their stock price, personally...

If you are interested, IBM does have a Linux POS software solution called IRES, based on Novell Linux. Check out http://www-03.ibm.com/products/retail/products/software/ires/ for more information. Good luck!

The POWER6 architecture is an incremental improvement on an architecture which dates back to the mid '90s

Early '90's, as in 1990.

You can see some evidence of the fact that IBM are trying to lower costs by sharing a lot of the design between the two lines though from certain new additions to the POWER instruction set, such as hardware support for Binary Coded Decimals (useful in high-throughput financial systems and present in the mainframe line since the 1401 and 700-series, which preceded System/360).

...and also present in the AS/400^WiSeries^WSystem i, which currently uses POWER processors, although the BCD stuff in POWER6 is probably as much for IEEE 754r floating-point decimal as for any commercial fixed-point BCD stuff.

The POWER6 architecture is an incremental improvement on an architecture which dates back to the mid '90s

Early '90's, as in 1990.

You can see some evidence of the fact that IBM are trying to lower costs by sharing a lot of the design between the two lines though from certain new additions to the POWER instruction set, such as hardware support for Binary Coded Decimals (useful in high-throughput financial systems and present in the mainframe line since the 1401 and 700-series, which preceded System/360).

...and also present in the AS/400^WiSeries^WSystem i, which currently uses POWER processors, although the BCD stuff in POWER6 is probably as much for IEEE 754r floating-point decimal as for any commercial fixed-point BCD stuff.

Nah, just a bunch of dual core chips. Take a look at the IBM Journal of Research and Development which has a lot of nice detail. Look at Vol. 48, No. 3/4 and Vol. 51, No. 1/2.

...If a processor is working on a particular piece of a problem and the data required to solve this will be made available by some other processor located away from the processor, I guess we are really talking wait times and not distance, am I right?

Depending on the network topology, it may be a question of actual machine-room distance. The BG/L, for instance, has a toroidal network, with each node connected to the 6 nearest nodes. There are four other networks in BlueGene, but the torus is used for most message-passing stuff. In that network, it's going to take longer to reach nodes that are physically farther away, unless you can wrap around the edges of the torus... see http://www.research.ibm.com/journal/rd/492/gara.html for more info.

For instance the Cell/BE processor allows C/C++ programmers to manage memory directly with the Memory Flow Controller to perform double-buffered asynchronous transfer of data between the main memory and the processor memory. Using Direct Memory Access, Cell users can achieve 98% of the peak performance on some applications: http://www-128.ibm.com/developerworks/power/library/pa-cellperf/ .

IBM has created a mouse cortical simulation:

http://domino.research.ibm.com/comm/pr.nsf/pages/rsc.bluegene_cognitive.html

They simulation is quire accurate, but it takes a Blue Gene computer to run it...the human brain has millions of such columns...

IBM's 1350 clusters are alive and well. Your choice, Xeon, Opteron or Power. RHEL or SLES?
http://www-03.ibm.com/systems/clusters/hardware/1350/

a) no mention of Bacon is made on the cycle collector page.
b) it labels the collector conservative, where Bacon's is accurate.
c) it mentions that your collector might fail to collect a garbage cycle, where by my recollection, Bacon's did not suffer this problem. Are you sure you're using Bacon's cycle collector? Perhaps these are simply the limitations of using C++.
d) why all the different collectors? Why not just use Efficient On-the-Fly Cycle Collection and be done with it? It's not quite as performant as a generational tracing collector, but at least your code won't suffer so many impedance mismatches, and thus be easier to maintain and extend.

IBM may be the best friend that the open source community has, but they are still interested in IBM's interests. A case in point is IBM's Rational Clearcase. If you look at the supported systems you will see only commercial Linux offerings. Redhat, Novell, and SUSE. Notably missing are CentOS, Ubuntu, and other free Linux in the x86/x86_64 platforms. If IBM is so pro-open source, why are they not supporting FOSS Linux on thier commercial packages?

IBM mainframes had multiple processors in 1994 it looks like: http://www.research.ibm.com/journal/rd/435/spainhower.html Although I didn't work with them until 2000 or so.

IBM has developed IDEMIX, a pseudonymous credential system. It work on the same principle and is going to be contributed to the Eclipse project as open source! http://www.zurich.ibm.com/security/idemix/ There is some white papers for those interested in the techno background.

• Tell the PHB's that the code you write will be human-comprehensible (suggesting maintanability), then create code with tags like <obfus:rangeScopingUpperOrLowerBoundDefiningIntermediateParameterIndex> so that only you (or someone with a few weeks to train in) can maintain your code
• Tell the PHB's that your code is interoperable while omitting the fact that it may take more logic to transform your XML so that it can "interoperate" than to generate it in the first place. More work for everyone!
• Perform the same type of magic as DBAs: "the information you want is in there, you just need to know how to get it out"
• More libraries = more bugs to work-around or fix

(The quadword instructions have 6 cycle latencies.)

If, by quadword instruction, you mean an instruction that operate
on 128-bit registers, this is incorrect.

First of all, 95% of the SPU instructions operate on 128-bit registers.
More than half of them have a latency of only 2 or 4 cycles. And the
vast majority of them have a throughput of 1 per cycle (ie. 0-cycle stall).
The only instructions with a 6-cycle stall are double-precision instructions
and 'fscrrd' (floating-point status control register read). See table B-1
in the Cell Broadband Engine Programming Handbook:
http://www.ibm.com/chips/techlib/techlib.nsf/techdocs/9F820A5FFA3ECE8C8725716A0062585F?Open&S_TACT=105AGX16&S_CMP=LP

This means for example that you can execute 1 'fm' per cycle (4-way
SIMD single-precision floating-point multiply), or 1 'and' per cycle
(4-way SIMD word add).

(The quadword instructions have 6 cycle latencies.)

If, by quadword instruction, you mean an instruction that operate on 128-bit registers, this is incorrect.

First of all, 95% of the SPU instructions operate on 128-bit registers. More than half of them have a latency of only 2 or 4 cycles. And the vast majority of them have a throughput of 1 per cycle (ie. 0-cycle stall). The only instructions with a 6-cycle stall are double-precision instructions and 'fscrrd' (floating-point status control register read). See table B-1 in the Cell Broadband Engine Programming Handbook: http://www.ibm.com/chips/techlib/techlib.nsf/techdocs/9F820A5FFA3ECE8C8725716A0062585F?Open&S_TACT=105AGX16&S_CMP=LP

This means for example that you can execute 1 'fm' per cycle (4-way SIMD single-precision floating-point multiply), or 1 'and' per cycle (4-way SIMD word add).

If you want a CELL supercomputer, IBM will happily sell you Cell/BE blade servers. The PS3 is a game console, the objective here is lower costs for the same hardware over a 6-7 year period.

As IBM says themselves in their paper Clarifying Misinformation on TCPA :

They have more reasons in that paper why their chip won't work with DRM.

A 6T SRAM cell consists of two cross-coupled inverters, and an access NMOS transistor on each side to connect the state nodes to bitlines. Writing is done by charging one bitline, and discharging the other. The devices should be sized such that the drive strengths allow the values on the bitlines to overwrite the values on the state nodes (i.e. win the drive fight). When you do a read, both bitlines are precharged, then the state nodes are connected to the bitlines by turning on the access transistors.

Now, when a read is happening, looking at the side of the bit cell that's holding, we see the bitline (at VDD), the access transistor which is on and looks like a resistor, the state node (initially at ground), and the pulldown transistor of the inverter which looks like a resistor connected to ground. This looks a lot like a voltage divider, and the state node will rise above ground. If it rises high enough, it could cause the bit cell to flip, corrupting its value. This is called a "read stability" problem, and for various reasons read stability tends to be a bigger problem at low operating voltages.

An 8T cell eliminates the read stability issue. It looks exactly the same as a 6T cell when it comes to writes, but reads occur on a separate bitline with a separate wordline. You add one transistor going between ground and a new node "foo", gated by the word line, and another transistor going between node "foo" and the read bitline, gated by one of the state nodes. A read no longer has any effect on the state node*, and won't corrupt the cell. You can see a schematic here (the right side - their "hit" signal would be the read wordline, and wl_ram would be the write wordline.

*ignoring Miller cap and other higher-order effects

Except none of what you just said stands up to basic scrutiny. First off, I can't find any evidence Black ever worked for IBM. He has been an author for decades, and wrote another book about the holocaust prior to the one focussing on IBM. Also, IBM themselves admit that the Nazis use hollerith machines in their crimes ( http://www-03.ibm.com/press/us/en/pressrelease/1388.wss ) their defence being that the German subsidiary was nationalised by the Nazis at the time.

Unlike Black, I suspect you are an employee of IBM

Oh, so I guess nobody else's shit stinks, right?

You could look at using tools such as IDS Scheer's Aris, or IBM's WebSphere Business Modeler. These allow you to capture and document processes in a way that goes well beyond anything you can knock up in Visio or a Wiki.

These companies will be only too happy to provide you with consultants who will help you use these tools within the context of a process modelling/documentation exercise.

All you need are deep pockets.

http://www-03.ibm.com/press/us/en/pressrelease/23370.wss The IBM press release does not specify any OS versions.

http://www-03.ibm.com/press/us/en/pressrelease/23370.wssThe actual release from IBM mentions NO versions.

"What is mystifying is that rather than hiring a team of HCI experts to smooth smooth over the more difficult concepts and eliminate places where the product was gratuitously arcane, IBM just slapped a slick looking cosmetic face lift on the product."

I hope I can help here: Installed user base.

If they change Notes too much, existing users will dislike it. You won't find many here that would believe that, but it's true.

I've seen this happen in real life, and can give an example - Domino Web Access (formerly known as iNotes).

Domino Web Access (DWA from hereon) is the webmail implementation of Notes - mail only, and runs in IE/Firefox. Because of limitations in the browser (it's base target is IE6 as it's a corporate product after all!), it can't assume a tabbed working space or the same kinds of controls.

Two specific examples were reading/composing emails, and selecting documents.

In Notes, new documents - for either reading or composing - open as a new tab. In DWA, they open as a new window. A very minor change, and technically quite understandable. But it had curious results.
Longtime Outlook users and light mail users liked it.
Longtime Notes users and heavy mail/app users hated it.

Why did the Notes users hate the new window per email? Because they were used to seeing all open windows as tabs in their Notes client. It made everything easy to navigate, as what was Notes stayed in Notes. But DWA starts throwing more and more windows open, making navigation painful when using other apps as well. And if Windows XP collapses the windows on the taskbar into one, now your DWA windows are confused with other apps you may be using in IE.
That's why I include heavy mail/app users as hating it. Whether they hated Notes or not, whether they preferred the look of DWA, they ALL wanted Notes back after less than two weeks of usage because of this.

By contrast, long-time Outlook users and light users were more likely to go for the rightmost, uppermost X on their screen and accidentally close the whole Notes client, losing everything open with obvious frustrations. (This was in the R6.5 days, before Notes 7 put a prompt up.)

Small change, big results.

Document selection is the other one that frustrated. Notes has an unusual selection system within views and folders - it makes it easier to select individual documents manually (you don't hold down CTRL, you click in the selection gutter or use the scroll keys and spacebar). But it makes it harder to select large swathes of documents as it doesn't support the CLICK --> SHIFT+CLICK method. (Until Notes R8, anyway.)

Of course, being implemented in a browser, DWA is the other way around. So those used to selecting apparently disparate (in terms of subject, sender or date) but actually related mails for foldering in Notes found DWA very frustrating. Whereas those coming from Outlook who were used to traditional selection found DWA far more to their liking.

(Funnily enough, the light/heavy divide went was the same here, too! Heavy users hated DWA, light users loved it. I'd like to think I had enough sample data to draw a conclusion, but I'm afraid that with only one implementation and less than two hundred users on this project, I'd rather say it's anecdotal than a hard fact...)

Notes has a rich interface which millions of people worldwide are used to. They're trained in it, or have picked up little tricks in it. If you ripped and replaced it wholesale with a "modern interface", then IBM would have serious problems selling the version after that due to the complaints.

They're aware of this high-wire that they have to balance on, and do employ HCI experts. Check out Mary Beth Raven's blog at http://www.ibm.com/developerworks/blogs/page/marybeth - she's the lead product designer for the UI of Notes right now, and agonises - in public, on that blog - about balancing new users versus old users.

Version 7 wasn't that much better than 6.5 in my opinion. I am currently running 7 and its still a lot of the things that people complain about (bloated, hard to use, slow, unintuitive). However I am very excited about the upcoming LN8 (upcoming in my company I mean - its already out from IBM) because of the completely revamped user interface.

Here is a link with some "whats new" information from IBM - http://www.ibm.com/developerworks/lotus/library/notes8-new/

http://www-1.ibm.com/support/docview.wss?uid=swg27009485

Linux versions supported:
# SUSE Linux Enterprise Desktop (SLED) 10 XGL
# RedHat Enterprise Linux (RHEL) 5 - Note: AIGLX and SELinux must be disabled

This just means if they find a bug they will fix it if they can reproduce on those platforms.

No need to target a VM, use PERL as the man says Using Inline in Perl

Where did they think that money was going to come from? That IBM would suddenly have that much extra money to throw around?

I don't have 2007 numbers yet, but here's IBM's reported profits from 2006:

Total Revenue -> $91.424 billion
Total Cost -> $53.129 billion
Gross Profit -> $38.295 billion

Yes, they have that much extra money to throw around.

Source: http://www.ibm.com/investor/financials/index.phtml

HUH?.....I guess Crazy Taco is more than just a nick for you.

You obviously have no idea all of the industries IBM is into. I am by no means an IBM fanboy but your assertions seem to be based on a very limited view of the company. TFA does not mention which arm of IBM this affects and I think that matters. TFA also does not say what the typical roles of these positions is. If this is for a desktop tech then a 15% reduction from 80K is pretty reasonable IMHO but if it's for a sysadmin it's not.

IBM dropped their PC business because PC's have been a commodity industry for years now and not at all profitable when you compare it to their "shrinking" mainframe business (Just one deployed mainframe can mean millions in contract support annually) not to mention their services arm (IBM Global Services or IGS) which last I heard was their most profitable enterprise (unconfirmed).

It's 1996, and I'm working at a university where the department IT guy is a rabid OS/2 fanatic. The whole department ran on Warp, but this brand new version of NT (4.0) has just come out with a Win95-like interface but decent internals, so the battle was on.

One day I wander down to the campus bookstore. They have copies of OS/2 in stock- the version with TCP/IP and a web browser was something like $200. Next to it was the development kit, in a plain box- $700.

On the other shelf is a copy of WinNT 4.0. $99. That $99 was the full version, and it included a full copy of Visual C++ as well.

IBM simply didn't care about the academic market at all. MS cares a *lot*- they learned from Apple that if you get people hooked earlier they are stuck with you for life.

If you're using systems with TCPA chips, then check out this overview and IBM's examples.

This sounds like Google is creating a ManyEyes site for the scientist set. http://services.alphaworks.ibm.com/manyeyes/app it's a lot of fun, but I don't see the Google version making neat things like word trees of the Grimm Fairy Tales like I did here: http://services.alphaworks.ibm.com/manyeyes/view/SmAgULsOtha65G-s4kxXL2-

This sounds like Google is creating a ManyEyes site for the scientist set. http://services.alphaworks.ibm.com/manyeyes/app it's a lot of fun, but I don't see the Google version making neat things like word trees of the Grimm Fairy Tales like I did here: http://services.alphaworks.ibm.com/manyeyes/view/SmAgULsOtha65G-s4kxXL2-

Relativity has a very robust tool for doing legacy code analysis:
http://www.regdeveloper.co.uk/2006/08/23/code_eam/
http://www.relativity.com/pages/home.asp
(also sold by IBM ftp://ftp.software.ibm.com/software/websphere/awdtools/atw/library/Analyzing_Programs.pdf)

Visustin is also worth looking at:
http://www.aivosto.com/visustin.html

http://www.fatesoft.com/s2f/

Clearly you don't write (or at least read source for) applications of any substance as that would be mildly described as tedious if not impossible.

One of the best ways to understand code is to do so visually with the software equivalent of blueprints. UML is generally considered a very capable way of modeling/communicating both static structures and dynamic behavior of software. There exist any number of tools that are capable of reverse-engineering existing source into UML. Two tools that I consider to be more capable than others are IBM's Rational Rose, and No Magic's MagicDraw. If commercial products aren't a possibility there are likely a number of open-source/free tools--though likely of lesser ability--available. A Google search on "reverse engineering UML" should point you at some.

Aww, I can't help myself. I keep posting this story whenever chess comes up. It's a short story he did once for putting an extremely shot story on a post card. Being slashdot, the majority have probably read it, but for the few it's a good quick read:

http://www.research.ibm.com/deepblue/learn/html/e.8.2.shtml

This story deals with there being more unique chess games than atoms in the universe.

And, funny, I just watched searching for bobby fischer last week. It's one of those kinds of movies I watch every couple of years. Just a classic.

Sometimes tools like Rational Rose or Enterprise Architect are successful at reading in the code an building a UML model that you can then attempt to parse through. I'm not familiar with the use of either, but I know it can be done, with mixed results depending on the size and complexity of the code being analyzed. Both tools are fairly expensive though, I believe.

• What is the nature of the "intellectual property" you are asserting protective ownership of? Is it copyright? Trademark? Patent? Some other unspecified fantasy form of IP (like SCO's undefinable "Linux stole it!" IP)?
• If it's not copyright, aren't you violating DMCA by invoking Digital Millennium Copyright Act takedown?
• If it's patent, which one and when was it filed?
• Ditto for trademark. Are you claiming, for instance, the board design is trademarked?
• If it's copyright, what elements? Are you claiming copyright over the phrase "Double letter score" or something? Are you claiming that Scrabulous has written instructions for the game which infringe your copyrighted written instructions for Scrabble?
• Or, most likely, aren't you just making this crap up and trying to bully on-line competition into giving up?

http://www-03.ibm.com/servers/eserver/about/cod/about/types.html

http://www-03.ibm.com/servers/eserver/about/cod/about/types.html

IBM's been doing this for years with some of their smaller servers http://www-03.ibm.com/systems/i/hardware/cod/index.html/

The cost in IT labor and lost productivity during the downtime that old methods need to add processing capacity can be a *lot* for servers hosting your important applications but its awfully expensive to pay upfront for enough power to keep up with ordering spikes during the Christmas buying season (for example) if that spikes way beyond your normal needs. Much cheaper to pay for only enough to handle your normal needs and then pay for the extra needed to handle oddball spikes only during the time you need it.

There's no way Ticketmaster's IT budget would agree to pre-pay for enough computing capacity to not bog down when the Hannah Montana tickets went on sale but if they could pay for just an hour of it ... much easier sale. Or remember the "performance" of Amazon's servers last Christmas when they put up their special sale items? If they could have just paid for a 24-hr keycode to enter the night you can bet the IT guys would have had a much easier time getting that in the budget.

Or as IBM puts it:
"Imagine you launch a dynamite new Web application for the holiday season, and it's getting more traffic than you expected. What do you do to avoid disruptions in service? You turn on available inactive processors and memory to handle every hit, then turn the extra capacity off when the application requires less capacity in the new year. You pay only for what you have activated.
Or say you tell your business analysts they now have access to all the company's business intelligence data. The danger is that, with your current processor configuration, increased demand could slow response times to a crawl. The solution? You activate reserve processing power to meet the new user demands without disrupting current operations."

The other beauty is that once the computer manufacturer has built in the ability to activate or inactivate processors and memory on the fly those same mechanisms make it natural to shuffle processors and memory between virtualized servers on the machine without restarting them.

And yes, it runs Linux.

Also in Unix/Linux Pseries and AS/400 Iseries servers. It is called Capacity On Demand :) CoD

Arguably, the PS3 is the most open console in history.

You do realise that the PS3 Cell processor is, from the ground up, built with Treacherous Computing, don't you? It's got a hardware DRM infection worse than any other consumer gadget.

The PS3 is more CLOSED than any console in history.

Sophia, Bulgaria was the home of the Dark Avenger one of the most notorious virus authors in history. He was quite active during the 80386/80486 time period. Some interesting reading about what is known of him can be found in these links: http://en.wikipedia.org/wiki/Dark_Avenger http://www.research.ibm.com/antivirus/SciPapers/Gordon/Avenger.html http://www.wired.com/wired/archive/5.11/heartof.html http://findarticles.com/p/articles/mi_m1511/is_n2_v14/ai_13381563/pg_9