Slashdot Mirror

Emperor Linux: $2750
Lenovo: $1,979.00

File systems became simpler over time. This is a GOOD THING AND THE ONLY WAY TO GO.

If you try to optimize too much, you end up with something like the IBM mainframe file systems from the 70s, which are still somewhat around.

Create a simple file, called a data set ? Sure, in TSO (what passes for a shell, more or less), you use the ALLOCATE command: http://publibz.boulder.ibm.com/cgi-bin/bookmgr_OS3 90/BOOKS/IKJ4C550/1.7.5?SHELF=&DT=20040721160158&C ASE=

Simple, isn't it ?

Forget complicated file systems, let the hardware handle speed. Ans possibly defragmentation.

Modern DASD architecture is almost completely hidden from the user. In the (good?) old days system software needed to interface closely with the DASD and needed to understand the hardware architecture to gain maximum performance from the devices (I know because I work on such systems within an IBM mainframe environment on airline systems which require extremely high speed data access).

Nowadays the disk 'address' of where the data actually resides is still couched in terms that appear to refer to the hardware itself but in 'serious' DASD subsystems (e.g. the IBM DS8000 enterprise storage systems )the actual way in which the hardware handles the data is masked from the operating system. Data for the same file is spread across many physical devices and some version of RAID is used for integrity.

The 4096 value for data 'chunks' has to do with the most efficient amount of data that can be transmitted down a DASD channel (between a host and storage in large systems or the bus in self-contained systems)

The idea of a 'file address' would cease to exist and it would be replaced by a generic 'data address' if it weren't for the in-built assumptions about data retrieval within all current Operating Systems.

There's the Fangs extension for Firefox (which doesn't actually speak). IBM Homepage Reader has a free evaluation download. JAWS screen reader has a demo version (if I remember correctly it expires after half an hour of use, then you have to reboot to use it again). Safari can be used as an aural browser. EMACSpeak can be used as an aural browser (in conjunction with EMACS-W3), and was the first (to my knowledge) to support aural CSS. However, it's only really any good if you have a hardware speech synthesiser - its software support depends on an old, obsolete library that is hard or impossible to get hold of these days. Opera has an aural mode.

Actually, that list surprises me somewhat. A couple of years ago, it was just JAWS, IBM, and EMACSpeak if you could get it to work. The past couple of years have really improved matters.

It depends on why you are testing really. If you are doing it for legal reasons, you're probably okay testing in just the aural browser that's most convenient for you. But if you are doing it for a genuine attempt at compatibility, you'll need to test in JAWS, it's the most popular by far.

It's quite a pain to use demo versions when you are only trying to make your websites more compatible with them. The best thing to do if you can't justify buying them is to develop a website so that it works without images, Javascript or CSS, and then test in the aural browsers, writing down everything that doesn't work quite right. Then fix up what you need to, and keep your notes to hand the next time you develop a site (i.e. write a "style guide" for your code).

I never said code for a specific browser -- I said don't use XHTML. I'm a supporter of valid code, and the Any Browser Campaign, but I don't currently support XHTML, because there is better cross platform support for HTML4 than there is for XHTML 1, because XHTML is not directly compatable with HTML

You can write pages that will degrade gracefully under both XHTML and HTML -- but you can't write a valid XHTML page that will parse cleanly in IE, XHTML is not backwards compatable with HTML, due to the need of an XML declaration before the HTML doctype.

To get XHTML to work in browsers that don't directly support XHTML, you'll have to serve it as text/html, which can then cause problems which correctly support XHTML. more details at wikipedia.

You get better overall support in browsers by coding to HTML4.01 -- there are few advantages to the user to using XHTML 1.0 (it's easier on the browser code, that's about it). That's not true with XHTML 2, which offers better alternative text for images and objects, as well as more easily arranged sections and dramatically updated Web Forms, but I have no idea when we'll get browser support for it. (hell, we don't even have good support for CSS2, much less CSS3 in most browsers)

You mean one of these, a T60/p?

I'll say right away that I don't have much experience with data centers, but it sounds like IBM's Tivoli software might be what you're looking for, except for the price factor. There's a whole load of inventory, monitoring (hardware and software), and event correlation / notification services available for the Tivoli Framework; there's no problem running it on VMWare systems, and it scales very well. I was just wondering if there are extra tools that a data center would need that Tivoli doesn't supply, or if you knew at all.

Then the dumbass news site is right. Take a look at this, or if you don't believe Google go here and take a look at the address.

I suspect that grandparent is one of the people who is convinced that TCPA == DRM. The fact is that Linux already has TCPA support (and, yes, you can just leave the checkbox unticked if it really bothers you). It's used for security, not DRM. TCPA is really just public-key encryption on a chip. Yes, that obviously could be used for DRM, but IBM (who contributed the TCPA support to Linux) has already published some whitepapers which, among other things, point out just how ineffective that would be and strongly hinting at some attack vectors--if you have access to the machine. (I can think of some more attack vectors, but they rely on having full control of the OS--fortunately, I do, thanks to IBM and Linus.)

Anyway, a bunch of the "TCPA is evil, it has no other purpose than DRM" crowd are running around shouting about how Linux is obviously about to require DRM--how else can they explain the TCPA support already there? Of course, the idea that switching to BSD will make you safe is quite silly--not only could TCPA support be added to BSD, but there's no guarantee that you'd get the source if that did happen! (Not that I have anything against BSD; I use it on a regular basis, and I'm a big fan, even though I generally prefer GNU userspace.)

1. 
   "Blurry"

   
   Most CRT's have circuitry that adjust for the aging process.
   
   "Hot"

   
   Most devices that generate work, generate heat. Including LCDs. Anyway I'm in front of a CRT and it doesn't generate as much heat as you think.
   
   "electricity-hungry"

   
   Not as much as you think when in a non-sleep mode. And even less sleeping.
   
   "large dangerous vacuum on your desk in front of your eyes"

   
   I can tell that you've never looked inside a modern CRT. One that vacuum is in a special glass envelope. Two most have a metal shield backing.
   
   "while sitting you right in the direction of an electron gun while firing x-rays out the back."

   
   Special glass, and metal shield. Plus soft X-rays aren't going to be coming out the back. (hint: look up how X-rays are generated). And last you get greater amounts of ionized radiaton from other sources.
   

I could take some pot-shots at LCDs.

I guess the PS3 HDD with Linux was true...
Not necessarily. AFAIK, IBM and friends have had Linux running on a cell for some time. And they plan on selling cell-based machines outside of the PS3. A quick google leads to this page : http://www-128.ibm.com/developerworks/power/librar y/pa-expert4/

Not to disagree that hardware and software go hand in hand with services, but software is not an afterthought. According to this : http://www.ibm.com/annualreport/2005/cfs_earnings. shtml they made quite a bit on software in 2005. And the margin on it was enormous.

For the ultimate in scalibility and reliability, you can run Lotus Domino on a zSeries mainframe.

I have to agree - Notes 7 is vastly improved from version 5.

IBM z/OS most certainly is UNIX-ish - that's because it IS a UNIX! Look here for more information.

z/OS UNIX System Services is UNIX95, XPG4, and XOpen compliant. What's neat about it is that you get the reliability of a mainframe with the flexibility of a UNIX system. You can have your legacy mainframe applications talk to your modern POSIX-based applications.

Disclaimer - I work for IBM, specifically within the USS product. That doesn't stop me from thinking it's a nifty product though :)

I wish I could mod that up. Obviously I missed the references in the original article. Thanks.

I wish I could mod that up. Obviously I missed the references in the original article. Thanks.

Not to pick, but being Unix-centric is not necessary. IBM's biggest systems run variations of the (old,proven,legacy,venerable) Z/OS http://www-03.ibm.com/servers/eserver/zseries/zos/ , and their iSeries (formerly AS/400) run i5/OS, http://www-03.ibm.com/servers/eserver/iseries/soft ware/os/i5os.html/ the successor to OS/400, neither of which is remotely unix-ish. I'll admit that they can run Unix, that the RS/6000s under AIX or Linux/PPC are unicies, or that at least virtual machines running under the primary OS can run Unix, but Unix-compatibility per-se isn't what Microsoft needs to compete against IBM.

What they need to compete is the high level of handholding, the extensive uptimes, and the absolute reliability and throughput of those IBM OSes. Microsoft will probably make inroads into the small-business market, and the edges of the corporation, but it's going to take more than just new software to displace IBM from the truly big-iron apps. Personally, I think that Sun, HP, and RedHat should be more concerned, as this will threaten the midrange server market.

Not to pick, but being Unix-centric is not necessary. IBM's biggest systems run variations of the (old,proven,legacy,venerable) Z/OS http://www-03.ibm.com/servers/eserver/zseries/zos/ , and their iSeries (formerly AS/400) run i5/OS, http://www-03.ibm.com/servers/eserver/iseries/soft ware/os/i5os.html/ the successor to OS/400, neither of which is remotely unix-ish. I'll admit that they can run Unix, that the RS/6000s under AIX or Linux/PPC are unicies, or that at least virtual machines running under the primary OS can run Unix, but Unix-compatibility per-se isn't what Microsoft needs to compete against IBM.

What they need to compete is the high level of handholding, the extensive uptimes, and the absolute reliability and throughput of those IBM OSes. Microsoft will probably make inroads into the small-business market, and the edges of the corporation, but it's going to take more than just new software to displace IBM from the truly big-iron apps. Personally, I think that Sun, HP, and RedHat should be more concerned, as this will threaten the midrange server market.

It's been quite some time, now since I first read about IBM's interest in PHP. I believe that this posting is further evidence that IBM intends on taping into the huge PHP development communities, allow IBM to quickly make up lost ground to Microsoft.

IBM's interest in PHP probably says as much about the established development communities as it does the language itself. I'm far from an expert on deploying technology globally, but I suspect that one of the biggest obstacles to overcome is finding qualified people to support it.

It might be a good idea for PHP developers to start looking at RUP. Learning formal software development processes can only help anyone looking the take their PHP skills to the corporate world.

In fact there are two articles on DeveloperWorks that I wrote last fall that describe how to use AspectJ to do more advanced performance monitoring (that are cited in the original article). See http://www-128.ibm.com/developerworks/java/library /j-aopwork10/ and http://www-128.ibm.com/developerworks/java/library /j-aopwork12/

The big benefits of using AspectJ are the ability to build more interesting coherent logic with a higher level language rather than (virtual) assembly language, and to have a well-document, accessible extension language for custom monitors.

Ron Bodkin

In fact there are two articles on DeveloperWorks that I wrote last fall that describe how to use AspectJ to do more advanced performance monitoring (that are cited in the original article). See http://www-128.ibm.com/developerworks/java/library /j-aopwork10/ and http://www-128.ibm.com/developerworks/java/library /j-aopwork12/

The big benefits of using AspectJ are the ability to build more interesting coherent logic with a higher level language rather than (virtual) assembly language, and to have a well-document, accessible extension language for custom monitors.

Ron Bodkin

Well, according to my buddy Mark, with BAMBIOS you'll be able to book x86 whatever you want (os/2, solaris, xp, etc). I'd guess we'll see a release within a week or so.

A hundred years of typing pools, and several hundred years of piano playing tell you how to sit and work, and it is no accident that proper typing posture is the same as proper piano posture.

I just Googled up an interesting site that discusses both issues.

Gnome's got a great library in GLib. I wrote a tutorial for IBM last year on the GLib collections; there are so many useful utilities and data structures in there. If you're writing a C app on Linux it's definitely worth a look, and if you're already using the GLib collections, take a look at that tutorial to see if you can optimize anything, like using g_list_prepend vs g_list_append.

And if it helps you, please buy my completely unrelated book!

Now combine that with coding/testing Zope 3 from a command line... and you've got some serious grue.

In that case you can buy a fire-resistant IBM ThinkPad. Or was that just an urban legend?

"I've seen a working version of this device and it seemed to have all the external features claimed in the article."

IBM had a working Linux watch years ago, and plenty of people saw it working, but it still became vaporware.

http://www.research.ibm.com/WearableComputing/linu xwatch/linuxwatch.html

The official statement, issued today through Reuters, is that the meeting will be an event to announce the future of the PlayStation side of Sony's business, which is particularly vague, although if we consider Koei's comments, it is highly likely there will be some mention of the the PlayStation 3. Next week also sees the Game Developer Conference 2006 taking place in the US, where it's known some SCEA developers will be speaking.

Sony's announcement:
Haha, tricked you! The truth is, there is no PS3. Sorry people, but you'll have to wait 6 more years for the PS4. Oh yeah, next week, we're having a conference about the PS4's capabilities

From the conference next week:
The PS4, AKA "Deus ex Machina," will be 100% compatible with 3d projectors, and will use IBM's hologram data storage as its storage medium. Please stay tuned for further updates.

...Amit Singh from IBM and kernelthread.com (slashdotted 16 times for excellent technical articles on various bits of internals of Apple hardware and Mac OS X) has his own legacy boot solution as well. From a rejected submission:

It appears that Amit Singh of IBM Almaden Research Center, of kernelthread.com and author of Mac OS X Internals, has devised a method to allow legacy, or BIOS-based, booting on Intel-based Macs, which they're calling "BAMBIOS". This means operating systems that currently only support legacy booting, such as many Linux distributions that don't yet support EFI, or things like Windows XP and the forthcoming Windows Vista (the 32-bit version of which will lack EFI support), will now be able to run on Intel-based Macs without modification (and completely legally). There is also another solution from "narf2006", described here and shown in this flickr set of photos. narf2006's solution is awaiting verification by Colin for the $12,000 pot. Time to get that MacBook Pro you've been waiting on for the best of both worlds, everyone...

So even if narf2006's solution isn't real, Amit's solution most certainly is, since he has a great deal of credibility. One way or another, we'll all be able to boot Windows directly on our Intel-based Macs.

This will be great news for people interested in Windows gaming on an Intel-based Mac (who really need the direct video access) and/or people who just want to do it NOW; however, a virtualization solution running under Mac OS X, such as VMware or Parallels, will be the real holy grail for most users. Most people don't want/need/care about the highest graphics and I/O performance; just the ability to run Windows side-by-side with Mac OS X at a speed that is more than usable, and to also have some capability to seamlessly share things like clipboards and files between the environments (as a nice VM environment would most certainly do). Not to mention not having to reboot.

In any case, even dual booting will be a welcome capability. It remains to be seen how convoluted the process is...

Also, I just spoke with Colin Nederkoorn (the guy running the contest) moments ago, and narf2006's solution has NOT been submitted to him yet. He said that narf2006 said he's "cleaning it up" and will be submitting it "later this week". So, no one, including Colin, has actually seen this solution working yet. Also, he apparently hasn't been in communication with Amit on the BAMBIOS solution as yet...

The issue you highlight is one of implementation of a practice, not within the practice itself. ITIL-based Service Management practices bring a high degree of process management and process maturity to an IT organization when implemented correctly. I would strongly caution against denegrating a product or practice when in actuality the problems lie elsewhere.

Also note that ITIL bills itself as a best-practice theory; think of it as the "logical" structure, not necessarily the "physical" structure. There are plenty of large IT companies that can work with your organization on successfully implementing an ITIL-based service management process framework, along with sophisticated products to back their processes up; Computer Associates (now CA) and IBM are two of the most prominent ones that come to mind.

I just took a trip to dell.com to check out, and they dont even offer 1000w for 1U servers.
Their dual Xeon 1U servers have 550W, with the option of dual 550W.

Another one here
ftp://ftp.software.ibm.com/common/ssi/rep_sp/n/XSD 00361USEN/XSD00361USEN.PDF
IBM x server 1U, dual xeon.
Maximum offered PSU: dual 585W.

I think you generalisations "dual 1000W, some dual 750W, NEVER seen a brand 500W" are full of shit.

Our apologies The IBM developerWorks Web site is currently under maintenance. Please try again later. Thank you.

Error: 500 Internal Server Error Server CoralWebPrx/0.1.16 (See http://coralcdn.org/) at 216.165.109.81:8090

I think that it was this last thing, the Federation interconnect, that they were pushing the data over in this test, since it forms the backbone of the machine...

The Federation switch is just the last incarnation of IBM's SP switch, which was the high-speed, low-latency, redundant, leaps-tall-buildings-in-a-single-bound, interconnect between the nodes of their Scalable Processor (SP) systems. Said systems were the basis for a bunch of the "we're at the top of the Supercomputing 500 list" IBM systems in the 90s, and probably into the current decade.

SPs were useful for highly parallel scientific processing, and were also great for some commercial applications like Oracle parallel (when you could get it to stay up). The high-speed switch beat the crap out of hooking systems together with Ethernet...

Originally, SP nodes were modified RS/6000 systems; they had the switch hardware tied very closely to the CPU/memory bus, and therefore had pretty fast data transfer around the system. Later, IBM started building cards to put into more normal HW, and (here's were I stopped playing with them) started to be able to connect other systems into a SP complex.

SPs used proprietary racks (called "frames") that had power distribution & the switch HW in the bottom, and up to 16 nodes in the frame. Part of the switch connections could go to other frames, so you could hook a bunch of them together. IBM officially supported some number of nodes (128? 256?) but you could go higher; that's what some of the high-end stuff in Sandia, etc., was.

Each node had up to 4 paths through the switch to get to any other node, so you had a lot of bandwidth on a continuous basis. You could run either TCP/IP or some other (don't remember what it was called) protocol over the switch.

GPFS was originally designed to handle multi-media files (video). When I used it (5 years ago) it was an additional layer that allowed SP nodes to access data from disks attached to other nodes with only a 7-instruction penalty. All of the communication between nodes went over the switch, so you really didn't care which node the data actually was on.

Of course all of this is from ancient memory, but it's probably close, and I don't feel like doing the research (this is /. after all...) http://www.ibm.com/ if you want.

or maybe you meant just opterons

From the articles I've read, this was accomplished using (some subset of) ASC Purple, which is full of a lot of either custom or IBM-proprietary stuff (or else stuff that nobody but IBM seems to be using).

According to the published/unclassified spec sheet:

"Purple has 2 million gigabytes of storage from more than 11,000 Serial ATA and Fibre Channel disks. ... Each login node has eight 10-gigabytes-per-second network connections for parallel file transfer protocol and two 1-gigabyte-per-second network connections for network file systems and secure shell protocol. The system has a three-stage 1,536 port dual plane Federation switch interconnect ..."

I think that it was this last thing, the Federation interconnect, that they were pushing the data over in this test, since it forms the backbone of the machine and links the storage nodes to the login node controllers, which then connect to the login nodes themselves (of which there are apparently over 1,400 of, according to this). I couldn't find much information on Federation, as it seems to only be used in a few systems, of which Purple is the most notable. One reference I found seems to put it at 1.49 GB/sec (11.92 Gbit/s) bandwidth, although it's not clear if that's "dual plane" Federation or not. 4X SDR Infiniband is around 10 Gbit/sec, IIRC, so Federation's a little faster.

Federation was the project name. HPS is the official product name. It was originally supposed to be a NUMA interconnect, but IBM changed their mind on NUMA. IBM's performance report is here:

http://www-03.ibm.com/servers/eserver/pseries/hard ware/whitepapers/pseries_hps_perf.pdf

The theoretical max banwidth is 2 gigabytes/sec for send and 2 gigabytes/sec for receive for each adapter, simultaneously. With the big Squadrons box, 16 adapters can be used to achieve a max of 64 gigabytes/sec/machine. Federation GA 1 was a rather bulky zero copy protocol stack. GA 2 was a return to the FIFO packet mode that was the basis of SP2 (colony/corsair) (yes, to save our asses because the zero copy architecture was horrible wrt latency). GA 3 added RDMA capabilities to the FIFO packet mode stack, which provided low latency __and__ high bandwidth.

"dual plane" means that each node is connected to 2 different Federation networks. They built the machine with 2 separate networks to reduce the number of hops that it takes to cross the network. Even with 2 planes it still can take 3 hops (they seem to always be chopping the machine up...).

There is more info here:

http://www.redbooks.ibm.com/redbooks/pdfs/sg246978 .pdf

Hitachi's SR11000 machines use POWER5, AIX5L, and Federation.

http://www.hitachi.co.jp/Prod/comp/hpc/SR_e/11ktop _e.html

From the articles I've read, this was accomplished using (some subset of) ASC Purple, which is full of a lot of either custom or IBM-proprietary stuff (or else stuff that nobody but IBM seems to be using).

According to the published/unclassified spec sheet:

"Purple has 2 million gigabytes of storage from more than 11,000 Serial ATA and Fibre Channel disks. ... Each login node has eight 10-gigabytes-per-second network connections for parallel file transfer protocol and two 1-gigabyte-per-second network connections for network file systems and secure shell protocol. The system has a three-stage 1,536 port dual plane Federation switch interconnect ..."

I think that it was this last thing, the Federation interconnect, that they were pushing the data over in this test, since it forms the backbone of the machine and links the storage nodes to the login node controllers, which then connect to the login nodes themselves (of which there are apparently over 1,400 of, according to this). I couldn't find much information on Federation, as it seems to only be used in a few systems, of which Purple is the most notable. One reference I found seems to put it at 1.49 GB/sec (11.92 Gbit/s) bandwidth, although it's not clear if that's "dual plane" Federation or not. 4X SDR Infiniband is around 10 Gbit/sec, IIRC, so Federation's a little faster.

Federation was the project name. HPS is the official product name. It was originally supposed to be a NUMA interconnect, but IBM changed their mind on NUMA. IBM's performance report is here:

http://www-03.ibm.com/servers/eserver/pseries/hard ware/whitepapers/pseries_hps_perf.pdf

The theoretical max banwidth is 2 gigabytes/sec for send and 2 gigabytes/sec for receive for each adapter, simultaneously. With the big Squadrons box, 16 adapters can be used to achieve a max of 64 gigabytes/sec/machine. Federation GA 1 was a rather bulky zero copy protocol stack. GA 2 was a return to the FIFO packet mode that was the basis of SP2 (colony/corsair) (yes, to save our asses because the zero copy architecture was horrible wrt latency). GA 3 added RDMA capabilities to the FIFO packet mode stack, which provided low latency __and__ high bandwidth.

"dual plane" means that each node is connected to 2 different Federation networks. They built the machine with 2 separate networks to reduce the number of hops that it takes to cross the network. Even with 2 planes it still can take 3 hops (they seem to always be chopping the machine up...).

There is more info here:

http://www.redbooks.ibm.com/redbooks/pdfs/sg246978 .pdf

Hitachi's SR11000 machines use POWER5, AIX5L, and Federation.

http://www.hitachi.co.jp/Prod/comp/hpc/SR_e/11ktop _e.html

"I think that it was this last thing, the Federation interconnect, that they were pushing the data over in this test, since it forms the backbone of the machine and links the storage nodes to the login node controllers, which then connect to the login nodes themselves (of which there are apparently over 1,400 of, according to this). I couldn't find much information on Federation, as it seems to only be used in a few systems, of which Purple is the most notable. One reference I found seems to put it at 1.49 GB/sec (11.92 Gbit/s) bandwidth, although it's not clear if that's "dual plane" Federation or not. 4X SDR Infiniband is around 10 Gbit/sec, IIRC, so Federation's a little faster."

Did some research and found the following:
http://www.llnl.gov/asc/platforms/purple/configura tion.html
http://www.redbooks.ibm.com/abstracts/sg246978.htm l - Info on the switch used

It appears to be an IBM thing that is only used on these big ASC platforms. The other parts of the company are using InfiniBand quite a bit though.
-Ack

For that matter, how does it compare with Tivoli TotalStorage SAN Filesystem, which seems to be another shared-storage filesystem from IBM/Tivoli? Trying to read IBM's descriptions is an exercise in marketing-fluff cryptography.

Much better info here (pdf).

GPFS (apparently -- I know only of what I've learned in the last few hours) is available for Linux, from IBM, right now.

Some people further up in the discussion have warned however that it's not as stable on Linux as it is on AIX, which is really its native platform.

From IBM's page on GPFS:

"GPFS is available as:
        * GPFS for AIX 5L on POWER(TM)
        * GPFS for Linux on IBM AMD processor-based servers and
            IBM eServer® xSeries®
        * GPFS for Linux on POWER"

GPFS Whitepaper - http://www-03.ibm.com/servers/eserver/pseries/soft ware/whitepapers/gpfsprimer.pdf

"GPFS is a cluster file system providing normal application interfaces, and has been available on AIX® operating system-based clusters since 1998 and Linux operating system-based clusters since 2001. GPFS distinguishes itself from other cluster file systems by providing concurrent, high-speed file access to applications executing on multiple nodes in an AIX 5L cluster, a Linux cluster or a heterogeneous cluster of AIX 5L and Linux machines. The processors supporting this cluster may be a mixture of IBM System p5(TM), p5 and pSeries® machines, IBM BladeCenter(TM) or IBM xSeries® machines based on Intel® or AMD processors. GPFS supports the current releases of AIX 5L and selected releases of Red Hat and SUSE LINUX Enterprise Server distributions. See the GPFS FAQ1 for a current list of tested machines and also tested Linux distribution levels. It is possible to run GPFS on compatible machines from other hardware vendors, but you should contact your IBM sales representative for details.

GPFS for AIX 5L and GPFS for Linux are derived from the same programming source and differ principally in adapting to the different hardware and operating system environments. The functionality of the two products is identical. GPFS V2.3 allows AIX 5L and Linux nodes, including Linux nodes on different machine architectures, to exist in the same cluster with shared access to the same GPFS file system. A cluster is a managed collection of computers which are connected via a network and share access to storage. Storage may be shared directly using storage networking capabilities provided by a storage vendor or by using IBM supplied capabilities which simulate a storage area network (SAN) over an IP network.

GPFS V2.3 is enhanced over previous releases of GPFS by introducing the capability to share data between clusters. This means that a cluster with proper authority can mount and directly access data owned by another cluster. It is possible to create clusters which own no data and are created for the sole purpose of accessing data owned by other clusters. The data transport uses either GPFS SAN simulation capabilities over a general network or SAN extension hardware.

GPFS V2.3 also adds new facilities in support of disaster recovery, recoverability and scaling. See the product publications for details2."

GPFS Whitepaper - http://www-03.ibm.com/servers/eserver/pseries/soft ware/whitepapers/gpfsprimer.pdf

"GPFS is a cluster file system providing normal application interfaces, and has been available on AIX® operating system-based clusters since 1998 and Linux operating system-based clusters since 2001. GPFS distinguishes itself from other cluster file systems by providing concurrent, high-speed file access to applications executing on multiple nodes in an AIX 5L cluster, a Linux cluster or a heterogeneous cluster of AIX 5L and Linux machines. The processors supporting this cluster may be a mixture of IBM System p5(TM), p5 and pSeries® machines, IBM BladeCenter(TM) or IBM xSeries® machines based on Intel® or AMD processors. GPFS supports the current releases of AIX 5L and selected releases of Red Hat and SUSE LINUX Enterprise Server distributions. See the GPFS FAQ1 for a current list of tested machines and also tested Linux distribution levels. It is possible to run GPFS on compatible machines from other hardware vendors, but you should contact your IBM sales representative for details.

GPFS for AIX 5L and GPFS for Linux are derived from the same programming source and differ principally in adapting to the different hardware and operating system environments. The functionality of the two products is identical. GPFS V2.3 allows AIX 5L and Linux nodes, including Linux nodes on different machine architectures, to exist in the same cluster with shared access to the same GPFS file system. A cluster is a managed collection of computers which are connected via a network and share access to storage. Storage may be shared directly using storage networking capabilities provided by a storage vendor or by using IBM supplied capabilities which simulate a storage area network (SAN) over an IP network.

GPFS V2.3 is enhanced over previous releases of GPFS by introducing the capability to share data between clusters. This means that a cluster with proper authority can mount and directly access data owned by another cluster. It is possible to create clusters which own no data and are created for the sole purpose of accessing data owned by other clusters. The data transport uses either GPFS SAN simulation capabilities over a general network or SAN extension hardware.

GPFS V2.3 also adds new facilities in support of disaster recovery, recoverability and scaling. See the product publications for details2."

Furthermore, it was also more than publicly announced at Lotusphere 2006 in Orlando, along with the Mac OS/X Notes 7.x client - so yes, it's no big secret. Additionally, there's an FAQ you can consume referenced at the bottom of this post. Sadly, the Mac Notes client demo received a rousing cheer and sustained applause from the audience, while the Linux Notes client garnered only a murmur of appreciation. Maybe the audience was largely IT execs / CIO types... *grin*

IBM Workplace is separate from the Notes client altogether however certainly acts as another solid-performing client, and is predicted by market analysts to slowly replace Notes reaching a potential 45 to 50% of Domino Server client seats in 2009 (Radicati - Market Analysis Lotus Notes 2005-2009 (go buy this yourself, it's $2500 USD so I'm not sharing)). IMHO, IBM are demonstrating a move away from the language of 2005 that was kind of claiming that Workplace and Notes are all the same thing - confusion from the user community seems to have prompted this. Workplace is just another client --> IBM quotes, "Q. How does "Hannover" differ from previously discussed plans for Lotus Notes 8? A. In Lotus Notes and Domino 7, we enabled Lotus Notes applications to be surfaced within the IBM® Workplace Managed Client(TM)." (Source: Hannover FAQ as at bottom of this post). Additionally, IBM say that Hannover will consume more of the Workplace frameworks (and will possibly be eventually superseded / EOL'd).

Lastly, Hannover is not a Linux port, Hannover is simply the next version of Notes after 7.0 (probably 8.0) - and will include the usual Windows platform on the client side as well. Windows pulls nearly 80% of Domino server installs (anecdotal data from Gartner for 2005) by platform, with AIX and Solaris following at approx 7 and 6% respectively (IBM claim not to track this - I suspect because it's somewhat embarrasing).

(Other sources; http://searchdomino.techtarget.com/originalContent /0,289142,sid4_gci1098021,00.html and http://www-142.ibm.com/software/sw-lotus/products/ product4.nsf/wdocs/hannoverfaq)

You mean, like IBM's up-and-coming port of the next version of Lotus Notes ("Hannover") to Linux and MacOSX, as part of their Workplace suite?

You can read it on this IBM link: http://www-306.ibm.com/software/swnews/swnews.nsf/ n/nhan6dbjwg

But as to the second part, IBM does still sell "Workstations" which are known to the general public as "desktop computers" (note in that link that its easy to find and buy "Workstations" through IBM's site, but impossible to find laptops). It sold only its "personal computing" business to Lenovo, which includes the "Thinkpad" and the "ThinkCentre" lines of computers, not the "IntelliStation" line, which is aimed toward business computing.

Finally, even if IBM was purchasing computers (and now it must purchase laptops from a third-party vendor), since it buys 1000 in a shot, I'm sure it can include something in the contract along the lines of "the computers will be shipped with the image supplied by IBM." IBM would do this so it doesn't have to pay the extra cost of having Windows on its machines.

Lastly, you could be correct that this is just a lot of hot air--after all it's just some rumor that even IBM itself denies. All I was saying was that it was rash to say "'No one will upgrade to Vista' doesn't mean that new machines won't be bundled with Vista." It's technically true, but your interpretation doesn't make sense in this context--supposedly he said this in a private meeting with close employees, and the company denies it having been said, even. This is not some political statement that was said to make the masses happy.

For the longest time Lotus had available a SmartSuite, a reasonably full featured office suite for Windows. While SmartSuite was never a leader in its field -- usually playing third behind Microsoft and WordPerfect -- it included notable components such as the venerable Lotus 1-2-3 and Freelance Graphics. If IBM and Lotus are indeed switching internal and client desktops to OpenOffice.org, is there any reason not to open source SmartSuite, giving OOo (and other) developers access to components and techniques that just might may have value to them?

For the longest time Lotus had available a SmartSuite, a reasonably full featured office suite for Windows. While SmartSuite was never a leader in its field -- usually playing third behind Microsoft and WordPerfect -- it included notable components such as the venerable Lotus 1-2-3 and Freelance Graphics. If IBM and Lotus are indeed switching internal and client desktops to OpenOffice.org, is there any reason not to open source SmartSuite, giving OOo (and other) developers access to components and techniques that just might may have value to them?

or this

http://www.ibm.com/linux tells all.

IBM could do this now, there's a Security chip in most IBM laptops, Heck, the security tech used in TCPA was Developed by IBM

It would be suicide for them to drop MS, because everyone and their uncle will just switch to Dell, and they know it. He's probably talking about what they are doing internally at IBM, which I wouldn't be surprised if it was running AIX or some in house mainframe system.