Posts by David Halko

@Matt Bryant - can't read

"you said Slowaris was fastest all the time in all situations"

It seems you can't read, either.

It seems the only word which "fastest" appeared in the entire listing of comments from anyone was targeted specifically at number of sockets.

We now know that Matt Bryant:

- can't spell

- can't read

- can't quote

Is the HP deal with SUN tied to the rapid decline in Windows Server sales?

Considering the report for Q4 Server Sales...

http://www.theregister.co.uk/2009/02/25/idc_q4_server_numbers/page2.html

"In Q4, Windows server sales fell by 17.8 per cent to $4.8bn... Windows had a 35.3 per cent share of the $13.5bn revenue pie in the quarter, but Unix edged it out with $4.9bn in sales, down only 6.2 per cent from the year-ago quarter."

Is this a sign that HP is trying to keep from losing revenue in it's x86 lines at 17.8% and possibly shift those customers to the other side of the equation?

Adding another UNIX operating system to the mix may be an opportunity to isolate itself from the Windows server free-fall, since UNIX servers showed more stability in server sales.

Reduce That Voltagage More!!! @Steve & @Coward

If the system does not normally need more than 7.2 volts, then why is more voltage needed?

I have not seen a whole lot of CPU's and peripherals require 12 volts. Laptop hard drives will often run with only 5v now.

This being said, a properly designed system that does not require more volts does not need a battery that supplies more voltage - if the vendor built in a 12 volt battery, it is a waste, since 3.8 volts would just be used to generate heat through step-down circuitry!

Some time in the future, I would hope to see batteries using 3.6 volts with 24 Amps... to get some real life expectancy and further reduce heat production!

They are positioning themselves...

NetApp is positioning themselves well for possibly sourcing hardware (and software) from SUN, if their legal case goes badly.

A Network Appliance box with SUN inside would be a great combination, since:

- NetApp would be able to stop development on proprietary software technology that really does not buy them any advantage over Open Source Solaris & ZFS

- SUN is already producing platforms with PCIe

The bottnenecks mentioned are already solved by SUN, how about other vendors?

"More recently, researchers at Sandia National Laboratory released a paper showing that chips run out of gas at eight cores."

Eight cores on the SUN CoolThreads T1 and T2 processors scale near linearly with no sign of running "out of gas".

http://www.sun.com/processors/UltraSPARC-T2/features.xml

Solaris with SPARC scales near-linearly into dozens of cores on a single large machine instance (without clustering) - place those same mechanisms in a single piece of silicon and any common man can see that the research was done by stupid people.

Similar things was said about 4 processors per box where 8 processors per box would cause Windows NT to suffer performance degradation - but this was due to a poor OS design, not a poor hardware design.

"I/O and memory bandwidth issues keep the processors tapping their feet, waiting for data."

This is why a good system vendor will make sure there is ample memory and I/O bandwidth available. Moving from 4 access channels to 8 access channels, as people speculate that SUN is going to do, will resolve this issue.

http://www.c0t0d0s0.org/media/presentations/RockHotChips.pdf

Also, there is a wide-scale movement in SUN to move to more solid state hard disks in their storage will alleviate I/O issues. Even ZFS will support solid-state flash to augment disks, at the OS level, so software applications will not be required to tune specifically for new hardware. This seems to resolve this concern.

http://blogs.zdnet.com/Murphy/?p=1191

"Sun Microsystems already has eight cores per chip (with eight threads per core) with its 'Niagara' family of Sparc T series, and will boost that to 16 cores and 32 threads with its 'Rock' UltraSparc-RK processors, due later this year."

Will SUN demonstrate near-linear performance increments with "Niagra" CoolThreads T series as well as with with "Rock" RK series?

If the issues with scaling from 4 to 8 cores are isolated to the issues mentioned in this article, then it seems SUN has had them resolved.

I am sure that SUN is not the only hardware vendor who has these issues resolved.

We need some smarter people writing these article.

Anonymous: SPARC, Power, Itanium & Migration...

"So now my CIO and CEO have told me that our waiting is over; we will replace the E25Ks by year's end. What does Sun have to offer?"

Of Course, the SPARC64!

SUN has always resold other vendor CPU chips in servers, I don't understand why people are so concerned about Fujitsu. CoolThreads was made by another vendor before SUN bought them. There was the "e" and "i" families of the UltraSPARC III processors, which was a different family. TI manufactures chips for SUN. What is the problem here?

"why did sun just extend their OEM agreement with Fujitsu for four years? why are they still hawking SPARC64 gear?"

To provide a long life for the M series of servers from Fujitsu Partnership. SUN had kept the UltraSPARC III, IV, and IV+ around for A LONG TIME - even when SPARC64 was released! Keep in mind, Fujitsu was selling SPARC64 systems before SUN resold them and Fujitsu will most likely continue selling SPARC64 systems after SUN releases new processors. This is the beauty of OpenSystems!

"Why does Sun only want to talk to me about Fujitsu gear?"

It is not Fujitsu gear, it is gear partnered and sold by both companies. Fujitsu resells CoolThreads servers, by the way.

"What do you thing their acquisition of Glassfish, MySQL and others is all about?"

To smooth out the large peaks and valleys in the big iron sales. If you watched the latest quarterly announcement video, you can see how SUN is using software to smooth out quarterly results.

"Rock, and high-end gear in general, is inconsistent with this roadmap. It doesn't make any sense."

Once again, if you watched the quarterly announcement, you will see how hardware sales jump up and down every couple of quarters, and SUN is using the growing software portfolio in order to smooth out the profits & losses. The hardware business provides SUN the lion share of revenue while the software revenue is still only a sizable percentage.

The plan for ROCK is very consistent with SUN's spoken direction for the past decade.

"would anyone in their right mind bet the farm on version one of anything from the choas that is now Sun?"

This is why SUN has multiple lines of systems and Fujitsu will continue to be providing chips for the M series while whoever the chip manufacturer is will start to provide chips for the new systems.

There is more security with multiple vendors (i.e. SUN, Fujitsu, TI, Atmel, Cyprus, Solbourne, ROSS with SPARC) in an open architecture than a single vendor with a proprietary architecture (i.e. DEC with Alpha; HP with PA-RISC; SGI with MIPS; IBM with Power; Intel with itanium.)

"Itanium is already old technology and IBM's POWER6 chip, with it's dual cores and 5 GHz clock speed is a generation ahead"

SUN & Fujitsu (with SPARC) have been at 2, 4, and 8 cores for some time now as other vendors are still playing around with silicon for 4 cores. With multiple SPARC vendors moving to 16 cores on both their low, midrange, and high end lines - I hardly see IBM as a single vendor with as a "generation ahead".

IBM as a single vendor has a faster single thread, SPARC with multiple vendors has more throughput per server, x64 with two vendors has cheaper equipment.

Each architecture has their advantages & disadvantages and SUN plays well in every architecture category now (including IBM POWER, supplying OpenSolaris on POWER, as well as SPARC on IBM Blade chassis.)

Can Aways Get a Solaris 10 Based Storage System...

Hey, Look Folks!

http://www.sun.com/storagetek/open.jsp

It Supports Windows!

One Mile Passive RFID Read? BOGUS!

"the technology employs no encryption and can be read from distances of more than a mile"

Documents leverage passive RFID.

Better be a Photon Torpedo from the Starship Enterprise in a clear shot with a steel convex parabola on the other side of a passive RFID document to ensure the signal bounces back at the right distance! LOL!

Having worked in an industry with RFID usage, there are reasons to be concerned (unauthorized reading), but when I read junk science with rhetoric & impossible claims, it makes me want to just ignore the subject.

Now...

If only GM would not change this new concept car much.

The latest incarnations of the Volt look terrible, bring back the concept cars!

The Register seems to ignore the most critical usage...

While there is certainly commercial benefit to such a technology, the U.S. Government would not be directly funding such technology unless there was some strategic value to it's Constitutional Role (i.e. protection of the United States of America.)

It would seem that a "boomless supersonic" stealth air vehicle would be a terrific asset to the U.S. military... radar would not see it coming and people might not be able to clearly tell when it had already passed over.

This is a terrific technology for military usage!

This boffin is a buffoon

Hate to tell this boffin/baffoon - he forgot to look at the rest of the multi-processor world.

It seems SPARC has not been suffering the same problems as AMD or Intel, in regard to multiple cores. SUN has been producing 8 core CPU's with 8 crypto cores on a single die for years from 4, 6, and 8 cores with linear performance.

Moving beyond 8 core has to do with the memory bandwidth, memory interconnects, and cache coherency - design it right and move on.

Vendors who successfully do near-linear SMP performance into 100 physical processors (like SUN) are most likely able to succeed in moving into 16 and more cores per socket - since they licked the problem in the large scale and the only thing that is to be done is to stick the existing technology onto a piece of silicon.

You are wrong about what makes SUN money...

Think about what you said...

"what actually makes Sun money - or presumably would - would be delivering its xVM Server variant of the Xen hypervisor for the Galaxy machines, and demonstrating that xVM is cheaper and better than the ESX Server and Hyper-V options."

If xVM is cheaper and xVM costs SUN to develop, then it would cause SUN to LOSE MONEY!

Now, xVM's N1 based management seems to be compatible with VMWare and HyperV - so there seems to be very little concern with what else is marketed.

VMWare and HyperV has more visibility in the SMB market, so less confusion from a vendor may actually translate into more sales in the short term (i.e. pick between 2 rather well known options than 3 options where one is not so well known.)

Re: RE: Re: RE: Re: Re: Bill and Yet Another AC

It seems you missed some important facts, Matt.

"At the top are a whole lot of people without much IT knowledge called 'the board'." - yea, they buy IBM and HP... LOL!

About SPEC... "Strange then that Sun used to sprout them so often when SPARC had competitive figures. But now they've fallen off the performance map they're keen to avoid that type of comparison. What a surprise" - what a surprise that SUN led in 1, 2, and 4 CPU socket SPEC performance over the past 2 years and you didn't have a clue! What a trip!!!

"Sun failed to diversify and failed to innovate, which is why it is now so deep in the doodoo" - that is why they sell & support M$ Windoze, Linux, Solaris, SPARC, Intel, AMD? LOL!

"but the difference is hp and IBM work with the community an make money from Linux" - that is why NFS, JAVA, OpenOffice, cluster file system, etc. all came from SUN and SUN re-sells Linux servers? ha ha ha ha ha!

"open source OpenVMS, arguably a much better product than Slowaris" - I bet Solaris will fork processes faster than OpenVMS! What a trip!

You're funny, Matt!!!

HA HA HA HA!

If you are not free to earn money and save money - you are not free

Money is just a way for two people to agree upon a portable way to value someone's perceived work.

If you are forced to work for someone and not receive compensation, you are a slave.

If you are forced to give up your compensation that was saved from your work, you have been stolen from.

Lack of capitalism is a denial of human rights - it is called slavery... and human slaves denied are victims of thievery.

Virtualization, VirtualBox & SUN

The author writes, "Over the years, the server virtualization strategy at Sun Microsystems has been spotty"

SUN has been doing virtualization since 1997 - it is not spotty, as the author indicated.

The author writes, "the company has yet to explain where VirtualBox will fit into its server virtualization strategy, if at all, in the long term."

I am uncertain why the author does not understand how virtual box fits into SUN's virtualization offering.

The author may want to view this SUN web site for where VirtualBox fits in the overall portfolio:

http://www.sun.com/software/products/xvm/

or read the SUN web site how VirtualBox specifically fits in

http://www.sun.com/software/products/virtualbox/

or read the announcement back in February of 2008 of why VirtualBox was important

http://www.sun.com/software/innotek/

SUN's virtualization offering seems to be broader than any other hardware or software vendor...

- virtualize at the NIC level (CrossBow)

- virtualize at the file system level (ZFS)

- virtualize at the firmware (Dynamic Domains)

- virtualize at the HyperVisor (xVM Server & Logical Domains)

- virtualize at the OS (Resource Management, Solaris 9 Containers, Solaris 10 Zones)

- virtualize the client (VDI with SunRay)

- virtualize in any OS client (VirtualBox)

- virtualize the application in any OS client (X, Secure Gobal Desktop, Java VM)

- virtualization management/operations (xVM OpsCenter 2.0)

Are there holes? Other vendor holes are gaping in comparison!

Longevity of Solaris 10 is a beautiful thing

Dunstan Vavasour comments, "at a headline level they've been 'stuck at Solaris 10' for nearly four years now, and I find myself having to explain to customers why this is such a good thing."

Stuck? LOL!

Every operating system should have a life expectancy of greater than 5 years!

Considering how robust the feature set is in Solaris 10, one would wonder why we don't see these features in other operating systems with binary compatibility guarantees!

I don't know about you folks, but I love the fact that I have integration scripts that I build under Intel SVR4, which worked under Solaris 2.5.1 SPARC, and has moved painlessly to Solaris 10 SPARC - when I get bored, perhaps I will see if they work under Solaris 10 Intel...

About the only problem I have seen is that some software "flips out" at the tremendous quantity of resources available under Solaris 10, in comparison to Solaris 8 (i.e. file handles) and I had to "tune the box down" to stop offering so many resources (when all I do is upgrade the OS on the same hardware with the same software!)

I look forward to seeing some of the new OpenSolaris features in commercially supported Solaris. I do not look forward to the new packaging system... been making SVR4 packages for a decade under multiple operating systems - don't need another barrier for entry of software vendors.

SUN's Press Releases About Released Products

dedmonst says, "Are you serious? Sun must have talked about ZFS and Janus for over 2 years before shipping - how about Rock?"

I don't remember seeing any press release about Rock being released yet.

I also don't remember seeing SUN providing a press release about ZFS before it was released.

I don't remember ever seeing a press release about the release of Janus.

People in architecture need to plan for 1, 2, and 3 years out. This means that some level of information needs to be available.

SUN indicated that they will upgrade processors and processor uniboards for UltraSPARC chassis' until end of life, and they did that with revs of UltraSPARC III, UltraSPARC IV, and UltraSPARC IV+ --- performing press releases each time.

SUN had indicated that they will slot 4 core AMD processors into existing platforms once they were released and they did a press release when it happened.

If you are suggesting that other vendors are doing press releases for products that are not available or do not become available, then I would suggest that SUN is better then the other vendors.

There is a significant difference between an road map, educating the consumer while a product is in development, and announcement of a product release. Perhaps this is where the confusion is?

For the cost of the building, they could have a SUN Black Box and get the computers for free!

It is rather ironic - I am not sure how IBM convinced them to buy a building for their supercomputer with some unbenchmarked possible proprietary 8 core processors with a proprietary OS.

The last Power processor with a single socket got 53.2 SPEC CINT2006 Rate

If they would get a SUN Black Box...

http://www.sun.com/products/sunmd/s20/specifications.jsp

1 Black Box = 8 racks * 80 sockets = 640 Sockets

Need more capacity? Well, add another box and stack it.

Add a bunch of Open Source 8 core processors with a bunch of Open Source OS's (download OpenSolaris or Linux) and get the same performance.

http://www.sun.com/servers/coolthreads/t5140/

A SUN CoolThreads T2+ will give the university 85.5 SPEC CINT2006 Rate per socket. With 2 sockets per 1U, the university could get 160 SPEC CINT2006 Rate per U of rack space. I wonder if IBM will be able to keep up.

The university would get the same computing power rolled to them and usable today if they went with SUN.

What an awful idea to wait for something to be cooked up by IBM somewhere in 2010 that they could have had since April of 2008 from SUN at a lower cost.

Someone needs a better education!

To Anon Corward - "Oracle on a 12 core chip... "

An Anonymous Coward says, "Oracle does not recognize LDOMS for sub capacity pricing so Sun customers are forced to try to use zones"

http://www.oracle.com/corporate/pricing/partitioning.pdf

"Hard partitioning physically segments a server, by taking a single large server and separating it into distinct smaller systems. Each separated system acts as a physically independent, self-contained server, typically with its own CPUs, operating system, separate boot area, memory, input/output subsystem and network resources. "

"Examples of such partitioning type include:... This is not a comprehensive list of all the different types of technologies or resource allocation devices/programs that would fall into the category of Hard partitioning."

If capped zones are acceptable, LDOM is obviously acceptable, since CPU capacity is added a thread of a core at a time, has it's own OS, separate boot area, memory, separate I/O subsystem, and network resources.

All the requirements are clearly met, by the letter of the Oracle licensing law!

IBM has some serious compatibility issues?

I did not realize that IBM had such serious compatibility issues within it's own systems as to require something like this to go from (almost) 32 bits to 64 bits... or even from one version of an OS to another... what a nightmare...

I hope they enjoy their Transitive purchase!

Finally... a lower entry point CoolThreads

I have been waiting for a lower entry point on the CoolThreads.

This is really needed to compete with the now-in-existence quad-core from AMD and Intel.

A speed-bump is needed in the high-end, to pass the new 6 core Intel CPU's, as well.

It must be hard on the CoolThreads engineering staff, they finally have some competition that they have to deal with - I hope they perform well!

Running 10 Virtual Machines on a T2000 has been nice...

I have to tell you... running 10 virtual machines on a T2000 with nearly 0% load impact per virtual machine has been very nice... with ample per-thread performance...

I can only imagine what some of the newer CoolThreads machines will be like... this box will allow for an 800% increase in workload... http://www.sun.com/servers/coolthreads/t5440/

It is hard to image close to a dozen virtual OS instance on any other platform... CoolThreads have me sold...

Brian has a clue, the author does not.

Yep... if the patent office has time to invalidate all the patents, then SUN does not have to sue anyone and the NetApp claim will be withdrawn.

Government usually messes up, in regards to dealing with the private sector.

Seems like a great strategy - let the U.S. Government do their job, that they were deficient in doing during the time of the patent filing. Why pay lawyers to do the job of the government?

SUN is way too undervalued

SUN has great technology, the only issue is profitability.

StorageTek was a good move, to keep SUN in the DataCenter, for a future come-back. SPARC & Solaris 10 should be placed in those StorageTek units, to help increase the overall SPARC volume, to help decrease cost to the rest of the customer base.

SUN should buy NetApp, put SPARC & Solaris 10 in those patforms, and continue the trend with other appliance companies... once again, to increase SPARC volume and lower cost to the consumer.

Appliances worked for Apple...

Neat Idea...

This is no far stretch.

The IPhone can already plug into a television.

Just give me a Bluetooth Keyboard and Bluetooth Mouse for the iPhone!

(Make sure the Bluetooth Keyboard has a button to do zoom in and zoom out!)

to B. Siddons -- SUN is apparently not dying on the vine...

article -- But while the market grew 18.8 per cent compared to Q2 cy'07 only Sun (34.7 per cent) and NetApp (22.9 per cent) outgrew the market and gained share. The other six vendors all gave up share growing as follows:- Dell 18.3 percent - HDS 18.1 per cent - EMC 16 per cent - IBM 14.8 per cent - HP 3.4 per cent - FSC -2.8 per cent

If SUN is dying at the vine, they seem to have more sales than any other storage vendor! You can't gain in Market Share, out-grow the Market, and STILL be dying at the vine!

article -- Gartner reckon that Sun made stellar sales of its StorageTek 2000, 6000 and 9000 series products. Sun has also said that its Sun Fire x4500 'Thumper' hybrid storage/server grew billings 37 per cent year-over-year in that quarter. It retained its top ranking for Unix units shipped for the 19th successive quarter.

Once again, if SUN is shipping more UNIX boxes than any other vendor, they are not dying at the vine.

SUN is also increasing their Linux and Windows share - it is just a matter of time before they stabilize.

to PT - California Energy Crisis Clarification

PT expresses, "Everyone knows who and what caused the California energy crisis. Unregulated free market capitalism, red in tooth and claw."

California's clean air standards limited in-state electricity generation to natural gas because of its lower emissions. This is not free-market capitalism, because there is no choice, rather a dictation by the government. This was unrealistic regulation.

There was a drought in the Summer of 2000, reducing the amount of energy being supplied to California. This is a fact.

As the demand increased in areas, retail prices were artificially capped by the government, keeping the utilities from expanding production. This is also a fact.

California lawmakers capped electricity prices and required utility to buy electricity from spot markets when challenged with power deficiencies. Price caps are not free-market capitalism, it is dictatorship.

California had a weak electrical infrastructure and California lawmakers made it impossible for them to upgrade it. Let me explain.

Path 15 allows electricity to travel from North to South, had not been expanded for years, and became a major bottleneck point. Only 3,900 Megawatts could be sent over Path 15. This forced the purchasing of energy from only a few producers using limited natural gas resources. Limited choices and suppliers is not free-market, it is typical poor central planning. This bottleneck was known since the 1980's in California, yet the dumb lawmakers decided to price-cap and inhibit the utilities from fixing the problem. See http://www.wapa.gov/sn/ops/transmission/path15/factSheet.pdf

After struggling for years with the crisis and refusing to address the core issues (due to the unpopularity of diverse power plant construction in California by lobbying groups), Democrat Governor Gray Davis of California (in November 2003, right before being kicked out of office by the citizens of California) eventually bailed out the utilities using tax payer money, streamlined the application process for new power companies.

In 2004, Republican Governor of California Arnold Sch attended the commissioning ceremony of a new upgraded long haul electrical infrastructure to resolve the Path 15 problem.

California has experienced few problems since, but the resolution to the problem came too late (years of warnings, years of inaction by Gray.) Gray only had the guts to try to fix the problem when he lost his re-election bid and the green lobbies no longer had power over him.

- Had the diverse generators been available in the area which needed the power, natural gas prices would not have become a single bottleneck.

- Had Path 15 been upgraded while it was tightly regulated by the California government before partial deregulation, or been allowed to be upgraded via rate hikes during deregulation, there would not have been a problem.

- Had prices been allowed to rise for consuming electricity, usage would have been reduced normally, and rationing would not have been needed through brown-outs.

- The IEA says a 5% reduction in demand would have resulted in a lowering of prices by 50%, market forces (had they been allowed to function by the government) - see page 8 http://www.iea.org/textbase/speech/2003/phbilling.pdf

Diversity in power sources to eliminate the need to transfer power, proper capacity to manage power demands in transfer system, or complete deregulation to allow retail price increase to reduce energy consumption voluntarily would have kept the 2000-2001 crisis from happening in California.

It was a perfect storm in California - history shows all options were taken off the table by lawmakers and all citizens suffered.

It's not like someone with technical skill left...

SUN is an engineering company... an engineering company is a good thing for engineers in MySQL.

if the guy was involved in technical projects, I would be concerned... if the guy was a V.P. over Asia Pac & Japan and still doing coding, I would still not be concerned, since he was not doing his V.P. job.

I'm sure the departed will have a nice time in another startup company - and may even build another company for SUN to buy up later and profit again!

to RW - Government licensing programmers???

This is the most bizarre comment I have ever heard!

A Government licensed programmer will not make 0 defect software!!!

Government intervention actually CREATED the problem!!!

Various legislative efforts to force electronic voting before it was clearly ready for prime time. Whenever government legislates & mandates any kind of development for which technology is not adequately tested, defects like this happen.

The various government obviously did not properly UAT test the machines correctly. Because these machines were used in hundreds of thousands of locations, the problem was not caught ahead of time, one could safely say that NONE OF THE GOVERNMENT AGENCIES PROPERLY UAT TESTED THE MACHINES!!!

Finally, because Microsoft Windows was placed on the machines, the excuse of a possible virus being at the root of the cause was actually viable in the media. Virus ridden Microsoft Windows should NEVER had been allowed on those machines, just for the reason of widespread viruses, worms, and trojan horses. I am not surprised that the governments who were making the requirements for the vendors to make a machine actually did not take this into consideration.

Now, RW is saying the people who forced the change and failed in countless UAT testings is going to fix this by licensing programmers?

HA HA HA HA!!!

Ian - I think some have missed the point...

I don't think anyone has missed the point.

PC Vendors have been making machines which has the new capacity for years now. We are not talking about high-end mid-range systems, but regular old SUN PC platforms had this capacity for awhile.

Many of the other vendors got out of the 8 socket PC market (i.e. HP, Dell, IBM, NCR, etc.) when proprietary CPU vendors started announcing dual-core... and SUN entered into that market to take up the slack for the lack of desire to retain the engineering capability during the post-year-2000 financial challenge years.

Making memory chips a little bigger does not mean vendors should exit the market of trying to avoid I/O bottlenecks for applications.

For some vendors, like SUN, it means providing a PC with 512 Gig or 1 Terabyte of RAM while other vendors are providing 288 Gig.

Nice to have some pictures in this article!

I remember trying to read another roadmap article just a couple days ago - it is nice to have the images in this one!

Intel x86 looking like OpenSPARC CoolThreads; Single-Threaded Myth

Most software is already using multiple threads, from desktops to servers.

The last single-threaded OS was DOS with Windows 3.1... but there were even TSR's back then.

With the advent of Windows 95, spawn windows spawned new threads. Dozens of services require CPU time with NT. Menus are now being checked in the background for usage statistics & being being rebuilt reformed. Indexing is always happening in the background. Speech recognition leverages threading and lookups. People are playing music & video in the background on their PC's all the time. Video & Audio chatting is normal now, where those incoming and outgoing streams are constantly being compressed and uncompressed in the background. People use their computers while rip'ing CD's and burning CD's. People are using their computers while they are recording videos from their TV, as well!

People are always running with multiple applications on the screen at the same time, and those applications are normally running more than one thread.

Most servers heavily use multiple threads and processes. Often, there are hundreds of processes and thousands of threads (my 1 processor 6 core 24 threaded lightly loaded server runs 611 processes and 1786 threads, while my heavily loaded 8 processor 16 core sever runs 252 processes and 2568 threads.) Web services, databases, and applications leveraged by multiple users are heavily thread dependent.

More cores and threads allow these interactions to be more seamless, increase throughput, reduce overhead of context switches, better manage performance peaks during high user counts,

The single threaded operating system and application was a myth promoted by chip manufacturers who were behind-the-curve technologically... If it was not a myth, these chip-manufacturers would not have been investing trillions into multi-core and multi-threaded silicon.

to Anonymous Coward - Memory & RAID Clarifications

Anonymous Coward asks, "The memory is DDR2, which I'm not familiar with, so would be interested to know if the individual banks (ie pairs of dimms) have separate channels, or whether the entire memory array has only 2 channels...?"

There appears to be 2 channels per socket.

http://www.sun.com/servers/x64/x4250/specs.xml

"Up to 64 GB (16 x 4 GB) of PC2-5300 667 MHz ECC fully buffered DDR2 memory"

http://www.sun.com/servers/x64/x4250/server_architecture.pdf

"Using a Sun StorageTek SAS RAID Host Bus Adapter (HBA), internal SAS disk drives can be configured for RAID 0, 1, 1E, 10, 5, 5EE, 50, 6, and 60 —when mirroring is implemented, drives are also hot-swappable."

Of course, some of the specs double with the X4450 (i.e. 4 quad core processors in 2U high, oodles of drives, 128 Gig RAM), as noted in the above white paper. Intel has not traditionally made 4 CPU sockets scale very well, unfortunately.

"Multiple, independent Front Side Buses (FSBs) that act as high-bandwidth system

interconnects. The Intel Xeon 5000 Sequence processors support both 1066 MT/sec and 1333 MT/sec Front Side Buses enabling theoretical data transfer rates of 8.66 GB/sec (at 1066 MT/sec) or 10.5 GB/sec (at 1333 MT/sec). The 7000 Sequence supports Front Side Bus interconnects at 1066 MT/sec."

The diagram for the 5000 sequence Intel processors (used in the servers reviewed by this article) implemented by SUN indicates that there are actually 4 memory channels, two for each front side bus, each memory channel operating at half the bandwidth of the front side bus. You can see this on pages 15 & 18.

If there will be an I/O bandwidth problem, it will be at the Intel Front Side Bus level... and this would only be accentuated in the quad-processor quad-core platform, which will be released soon, due to the decreased speed of the Intel Front Side Bus in the 7000 sequence processors. You can see this on page 21.

to TimM - Windows does not dominate 64 bit

TimM says, "Let's be real here and recognise that Windows vastly dominates the desktop... The main issue with 64bit Windows is the lack of support by software and more importantly hardware manufacturers. Drivers really are the big problem."

Let's be real here and recognize that Windows does not dominate the 64 bit market.

You want 64 bit applications on a 64 bit operating system with 64 bit driver support.

Pick the operating system and platform to support your application requirement.

If your application does not support a real operating system, you picked the wrong application.

I have been using 64 bit operating systems for years and using 64 bit applications when we hit 32 bit memory boundaries (3.75 Gig under Solaris, 2 Gig under other operating systems.)

It is all about architecture.

Android on Java on an iPhone?

It seems funny... people are talking about Android, Google, and iPhone and not considering the obvious - Android on an iPhone.

It seems that the iPhone has a Java bytecode hardware engine built into it.

http://blogs.zdnet.com/Burnette/?p=338

Users can jailbreak the iPhone to get Java onto it.

http://blog.taragana.com/index.php/archive/how-to-install-compile-run-java-on-iphone/

The The Innaworks product, alcheMo for iPhone, could reasonably bring Android to the iPhone, without a jailbreak (although the following article does not mention Android as a targeted Java application.)

http://www.infoworld.com/article/08/04/24/Sun-continues-pursuit-of-Java-for-iPhone_1.html

With those new 3G iPhones at $199, it will not be long for the 1 generation iPhone to be available at a steal on eBay... they would make a good Android candidate!