Showing posts with label IBM. Show all posts
Showing posts with label IBM. Show all posts

Monday, November 4, 2019

Distributed Denial of Service, Amazon Cloud & Consequences

[Amazon Web Services Logo, Courtesy Amazon]

Distributed Denial of Service, Amazon Cloud & Consequences

Abstract

The US Military had been involved in advancing the art of computing infrastructure since the early days of computing. With many clouds built inside the Pentagon, a desire to standardize on an external cloud vendor was initiated. Unlike many contracts, where vendors were considered to compete with one another for a piece of the pie, this was a "live and let die" contract, for the whole proverbial pie, not just a slice. Many vendors & government proponents did not like this approach, but the proverbial "favoured son", who had a CIA contract, approved. This is that son's story.


Problems of Very Few Large Customers

Very few large customers create distortions in the market.
  1. Many understand that consolidate smaller contracts into very few large contracts is unhealthy. Few very large single consumers, like the Military, create an environment where  suppliers will exit the business, if they can not win some business, since the number of buyers is too small, limiting possible suppliers in time of war.
  2. Some complain that personal disputes can get in the way of objective decision making, in large business transactions.
  3. Others warn that political partisanship can wreck otherwise potential terrific technology decisions.
  4. Many complain that only a few large contracts offer opportunity for corruption at many levels, because the stakes are so high for the huge entities trying to gain that business.
  5. In older days, mistakes by smaller suppliers gave opportunity for correction, before the next bid... but when very few bids are offered, fleeting opportunities require substantially deep pockets to survive a bid loss
  6. Fewer customer opportunities discourages innovation, since risk to be innovative may result in loss of an opportunity when a few RFP providers may be rigidly bound by restraints of older technology requests and discourages from higher costing newer technology opportunities
In the end, these logical issues may not have been the only realistic problems.


[Amazon Gift Card, Courtesy Amazon]

Amazon's Business to Lose

From the very beginning, Amazon's Jeff Bezos had his way in. Former Defense Secretary James Mattis, hired Washington DC Lobbyist Sally Donnelly, who formerly worked for Amazon, and the Pentagon was soon committed to moving all their data to the private cloud. The irony is that Bezos, who has a bitter disagreement with President Trump, now had a proverbial "ring in the nose" of President Trump's "second in command" with the Armed Forces, in 2017.

Amazon's Anthony DeMartino, a former deputy chief of staff in the secretary of defense’s office, who previously consulted for Amazon Web Services, was also extended a job at Amazon, after working through the RFP process.

Features of the Amazon Cloud, suspiciously looked like they were taylor written for Amazon, requesting features that only Amazon could offer. Competitors like Oracle had changed their whole business model, to redirect all corporate revenue into Cloud Computing, to even qualify for the $2 Billion in revenue requirement to be allowed to bid on the RFP! How did such requirements appear?

Amazon's Deap Ubhi left the AWS Cloud Division, to work at the Pentagon, to create the JEDI procurement contract, and later return to Amazon. Ubhi, a venture capitalist, worked as 1 of a 4 person team, to shape the JEDI procurement process, while in secret negotiations with Amazon to be re-hired for a future job. The Intercept further reminded us:
Under the Procurement Integrity Act, government officials who are “contacted by a [contract] bidder about non-federal employment” have two options: They must either report the contact and reject the offer of employment or promptly recuse themselves from any contract proceedings.
The Intercept also noted that Ubhi accepted a verbal offer from Amazon, for the purchase of one of his owned companies, during the time of his working on the Market Research that would eventually form the RFP.

A third DoD individual, tailoring the RFP, was also offered a job at Amazon, according to Oracle court filings, but this person was marked from the record.

At the highest & lowest levels, the JEDI contract appeared to be "Gift-Wrapped" for Amazon.

[Amazon CEO Jeff Bezos hosting Trump's Former Defense Secretary James Mattis at HQ, courtesy Twitter]

Amazon Navigating Troubled Waters

December 23, 2018, President Trump pushes out Secretary of Defense James Mattis after Mattis offered a resignation letter, effective February 2019.

January 24, 2019, Pentagon investigates Oracle concerns unfair practices by hiring Cloud Procurement Contract worker from Amazon.

April 11, 2019, Microsoft & Amazon become finalists in the JEDI cloud bidding, knocking out other competitors like Oracle & IBM.

June 28, 2019, Oracle Corporation files lawsuit against Federal Government for creating RFP rules which violate various Federal Laws, passed by Congress, to restrict corruption. Oracle also argued that three individuals, who tilted the process towards Amazon, who were effectively "paid off" by receiving jobs at Amazon.

July 12, 2019, Judge rules against Oracle in lawsuit over bid improprieties, leaving Microsoft & Amazon as finalists.

August 9, 2019, Newly appointed Secretary of Defense Mark Esper and was to complete "a series of thorough reviews of the technology" before the JEDI procurement is executed.

On August 29, 2019, the Pentagon awarded it's DEOS (Defense Enterprise Office Solutions) cloud contract, a 10-year, $7.6 billion, to Microsoft, based upon their 365 platform.

On October 22, 2019, Secretary of Defense Mark Esper withdrew from reviewing bids on the JEDI contract, due to his son being employed by one of the previous losing bidders.

Serendipity vs Spiral Death Syndrome

Serendipity is the occurrence and development of events by chance with a beneficial results. The opposite may be Spiral Death Syndrome, when an odd event may create a situation where catastrophic failure becomes unavoidable.

What happens when an issue, possibly out of the control of a bidder, becomes news during a vendor choice?

This may have occurred with Amazon AWS, in their recent bid for a government contract. Amazon pushed to have the Pentagon Clouds outsourced, at one level below The President and even had the rules written for an RFP, to favor a massive $10 Billion 10 year single contract agreement favoring them.

October 22, 2019, A Distributed Denial of Service (DDoS) hitsAmazon Web Services was hit by a Distributed Denial of Service attack, taking down users of Amazon AWS for hours. Oddly enough, it was a DNS attack, centered upon Amazon C3 storage objects. External vendors measured the outages to last 13 hours.

On October 25, 2019, the Pentagon awarded it's JEDI (Joint Enterprise Defense Infrastructure) cloud contract, a 10-year, $10 billion, to Microsoft. The Pentagon had over 500 separate clouds, to be unified under Microsoft, and it looks like Microsoft will do the work, with the help of smaller partners.

Conclusions:

Whether the final choice of the JEDI provider was Serendipitous for Microsoft, or the result of Spiral Death Syndrome for Amazon, is for the reader to decide. For this writer, the final stages of choosing a bidder, where the favoured bidder looks like they could have been manipulating the system at the highest & lowest levels of government, even having the final newly installed firewall [Mark Esper] torn down 3 days earlier, is an amazing journey. A 13 hour cloud outage seems to have been the final proverbial "nail in the coffin" for a skilled new bidder who was poised to become the ONLY cloud service provider to the U.S. Department of Defense.

(Full Disclosure: a single cloud outage for Pentagon Data, just before a pre-emptive nuclear attack on the United States & European Allies [under our nuclear umbrella], lasting 13 hours, could have not only been disastrous, but could have wiped out Western Civilization. Compartmentalization of data is critical for data security and the concept of a single cloud seems ill-baked, in the opinion of this writer.)

Monday, October 29, 2018

Oracle Linux on SPARC is dead? Oracle Linux at Risk?

Oracle Linux on SPARC is dead? Oracle Linux at Risk?

Oracle made substantial changes in their strategy last year, perhaps on a "Wim"... and now they seemed to bet wrong.

Oracle has long pinned some of it's engineered systems on a clone of Red Hat Linux. After purchasing Sun, they released storage servers based upon Solaris on Intel and left it's other engineered systems on their knock-off Linux OS. 

Oracle had to suffer through the successive Intel CPU fixes making each successive patch release slower or less secure. Now, the dominate Linux Vendor [Red Hat] that Oracle had been copying is being purchased by IBM. 

[SPARC logo, courtesy SPARC International]

SPARC Life


It appears that there is still a SPARC of life in the world's highest performing CPU architecture... and that life in Solaris. There has been a recent roadmap release [ie 2018-08] which is substantially the same as it's previous release some 5 months earlier.

[Oracle logo, courtesy Oracle Corporation]

Oracle SPARC

Oracle releases a new roadmap with an M8+ chip coming (ie 2018-03) and continues to design Oracle Solaris, for the distant future, for over a decade.

Oracle SPARC Solaris appears to be a steady ship, in turbulent seas.

[Fujitsu logo, courtest Fujitsu corporation]

Fujitsu SPARC

This seems to coincides with the Fujitsu roadmap [i.e. since last year!] It seems Fujitsu is designing the silicon for Oracle as they advance the Solaris Software layer. Fujitsu, a hardware provider supplying SPARC chips for Sun when SPARC was first created, continues to talk about new product coming [in 2018-02-03, 2018-03-15] - which is good news!

Fujitsu leaked [in 2018-07-06] is getting closer to releasing it's new Supercomputer architecture, not based upon SPARC, which probably means Fujitsu's Linux for SPARC will soon have no future.

Oracle Linux

There has been some speculation about Linux on SPARC from NetMgt. The last update of Oracle Linux on SPARC looks like Summer 2017. It appears to have stalled, possibly killed when Wim returned to the Oracle in November 2017.

Oracle's knock-off Linux is based  upon Red Hat Linux... which is now being purchased [in 2018-10-28] by arch-enemy competitor IBM... who competes in all Oracle's major spaces (i.e. Cloud, RISC servers, Intel Servers, Database, etc.)

Conclusions

NetMgt has been tracking Oracle Linux for some time, but it appears Oracle Linux on SPARC stalled last year. Oracle Linux on SPARC now appears dead on arrival. Fujitsu SPARC no longer has a need for Linux. Oracle's Linux, is now oddly in a strange risk place, under Intel whose CPU's get slower with every defect fix. Oracle SPARC Solaris continues to be the highest performing Vendor Architecture and OS combination - the "sun" continued to shine in the darkness of declining performance of competitors.

Thursday, April 17, 2014

Hardware: American Sell-Off with IBM and Google

[IBM Logo, courtesy IBM]

Abstract:
As the misguided U.S. economy continues to run up massive debt and continue massive trade deficit, the sell-off of U.S. High Technology assets continues to non-U.S. companies, fat with outsourcing cash. Lenovo, a Chinese company, continues their purchases in the United States of inventors of technologyu.
[Chinese glorifying revolution, courtesy, The Telegraph]
Chinese Lenovo Purchasing U.S. Hard Technology

Chinese global company Lenovo has been purchasing their way into the U.S. market through many technologies essentially invented in the United States. IBM seems to be the most significant seller.

[IBM PC, courtesy Wikipedia]
  • 2005-05-01 - PC Division acquired from IBM (PC's and ThinkPad Laptops)
    Chinese computer maker Lenovo has completed its $1.75 billion purchase of IBM’s personal computer division, creating the world’s third-largest PC maker, the company said Sunday. The deal — one of the biggest foreign acquisitions ever by a Chinese company
    [IBM Thinkpad, courtesy tecqcom]
  • 2006-04-10 - Lenovo makes break with the IBM brand (on PC's, not ThinkPad Laptops)
    Since Lenovo took over the IBM personal computer business on May 1, 2005, the company's advertising and marketing efforts have excluded IBM almost entirely. The four television spots that Lenovo ran during the Turin Winter Olympics, for example, never mentioned IBM at all. In fact, the only connection to the iconic brand is the IBM logo, which still adorns Lenovo's ThinkPad laptops.
  • 2013-01-07 - Lenovo to create ThinkPad-focused business unit to compete at the high end
    Lenovo is reorganizing its operations into two business groups... As part of the restructuring, it will create two new divisions, Lenovo Business Group and Think Business Group.The reorganization, which will be completed on April 1 [2013]
    [IBM Servers, courtesy Wikipedia]
  • 2014-01-23- Lenovo to buy IBM's x86 server business for $2.3bn (PC Servers)
    Lenovo and IBM announced on Thursday they have signed a definitive agreement that will see the Chinese hardware giant acquire the IBM's x86 server business for the tidy sum of $2.3bn, with approximately $2bn to be paid in cash and the balance in Lenovo stock.
    Adding to the PC business Lenovo acquired from IBM in 2005, Lenovo will take charge of IBM's System x, BladeCenter and Flex System blade servers and switches, x86-based Flex integrated systems, NeXtScale and iDataPlex servers and associated software, blade networking and maintenance operations.
    [Motorola Droid RAZR, courtesy Wikipedia]
  • 2014-01-29 - Motorola Cellphone Company acquired from Google (by Lenovo)
    Lenovo has signed a deal to buy the loss-making Motorola Mobility smartphone manufacturer for $2.91bn, but a switched-on Google is keeping the patents owned by the firm it gobbled two years ago for $12.5bn.
    "The acquisition of such an iconic brand, innovative product portfolio and incredibly talented global team will immediately make Lenovo a strong global competitor in smartphones," said Lenovo's CEO Yang Yuanqing. "We will immediately have the opportunity to become a strong global player in the fast-growing mobile space."
  • 2014-01-29 -  Lenovo splits into 4 groups after buying IBM's server business
    A few days after announcing its plan to buy IBM’s x86 server business, the Chinese company is dividing its operations into four business groups... enterprise products... developing a software ecosystem...PCs and mobile products. The changes go into effect on April 1 [2014]
Clearly, Lenovo has a vision for the U.S. Market and is executing upon it. How unfortunate that American companies such as IBM and Google see little value or possibility in domestic hardware innovation, moving into the future.
[HP Logo, courtesy eWeek]
Impacts in the U.S. Market

There is a great deal of uncertainty felt by partners and customers of IBM through such acquisitions. Previous attempts to leverage the IBM logo to help assure customers was performed, but with the latest purchase - competitors such as HP are seeing the a lot of noise.
  • 2014-04-11 - HP: Lenovo's buy of IBM x86 biz is bad, bad, bad...
    "Customers and partners are concerned. They are concerned about what the future will be for them – not only in the product but also in support and services," claimed the exec veep and GM of the Enterprise Group.
    HP has an internal migration programme to support customers with IBM servers as they decide to make the switch, he pointed out.
    But providing maintenance support is something that HP and other vendors already offer on third-party kit as standard.
HP was tried to consolidate all of their computing systems under Intel Itanium, before trying to shut them all down. HP also tried to sell off their PC business, but relented, possibly due to customer pressure. How conservative customers who would only buy IBM will respond in the U.S. to their favorite manufacturer leaving the industry may not be a difficult conclusion to reach, especially from companies like HP.
Concluding Thoughts:
The massive technology bleed from the United States is partially due to commoditization, but also due to the migration to Cloud and Appliances and value provided by Intel computing vendors becoming less significant with Intel shipping entire motherboards bundling CPU, Floating Point, Memory Management Units, Ethernet, and most recently Video. Cell phones appear to be drastically simplifying, as well. Perhaps there was nothing of value left for Intel or cell phone based manufacturers to do? Can Apple buck the trend?


Wednesday, December 26, 2012

Networking: 2012 December Update


Cisco to sell off Linksys division; Barclays to Find Buyer

DARPA to Create 100Gb Wireless Skynet

Ethernet Switch Sales Decline, SDN (Software Defined Networks) to Explode...

IBM Integrates Optics onto Silicon...

Tuesday, December 25, 2012

New Technology: 2012 December Update


SSD prices are low—and they'll get lower. MeRAM posed to supplant NAND flash memory...

IBM Integrates Optics onto Silicon...

Sun/Oracle receives patent 8,316,366 on Transactional Threading on November 2012... this came on the heels of a 2011 paper on formally verifying transactional memory on September 2011.

Silence has been from Sun/Oracle VLSI group on Proximity Communications, research funding is due to expire in 2013, is there a product in the future?

Samsung Spends $3.9bn on iPhone Chip Factory in Texas.

Texas Instruments to cut 517 OPAM Smartphone/Tablet Chip Manufacturing jobs in France.

AWS (Amazon Web Services) Hosting Server Retirement Notifications Wanting...

Microsoft Outlook 2013 Willfully Broken: Will Not Recognize .doc or .xls Files

Microsoft Windows 8: Hidden Backup & Clone Feature

Wednesday, November 14, 2012

IBM POWER 7+ Better Late than Never?

Anticipation of POWER 7+
In August 2011, when Oracle released the SPARC T4, IBM decided to release something, but it was not a CPU chip, but a roadmap... problem was - IBM POWER 7+ was late.

Courtesy, The Register
 Not only was it late, but apparently it would be VERY late. In a previous Network Management article:
Thank you for the IBM August 2011 POWER Roadmap tha the public marketplace has been begging for... did we miss the POWER7+ release??? A POWER 7 February 2010 launch would have POWER 7+ August 2011 launch (and today is August 31, so unless there is a launch in the next 23 hours, it looks late to me.)
IBM Flex p260 CPU Board, courtesy The Register
Disposition of POWER 7+
Just announced, via TPM at The Register, is the ability to provision an IBM p260 with new POWER 7+ processors!

IBM finally released POWER 7+, in one system, 15 months late. How underwhelming.

IBM POWER 7+ die image, courtsy The Register
Disposition of POWER 8
The question that no one is asking: Where is POWER 8, according to the IBM roadmap?

The POWER 8, according to the roadmap, should be released roughly February of 2013. Why release the POWER 7+ just 3 months shy of releasing the POWER 8? This suggests a problem for IBM.

The 2013 timeframe is roughly when Oracle suggested they may start releasing SPARC T5 platforms, ranging from 1 to 8 sockets. The POWER 8 surprisingly has a lot of common features with the SPARC T5 -the POWER 8 may have gone back to the drawing board to copy SPARC T5. How late is POWER 8?
Conclusions:
Whatever happened, something broke at IBM. For a piece of silicon to be delayed 15 months, IBM must have needed to bring POWER 7+ back to the proverbial drawing board. The big question is - did IBM need to bring POWER 8 back to the drawing board, as well?

Thursday, September 27, 2012

POWER Double Stuff vs SPARC Critical-Thread

[Sun UltraSPARC T3 16 core CPU Diagram]
Abstract:
Computing processor models differ in architecture from one company to another, each trying to gain an edge in the market over their competitors. Often, chip foundries will attempt radical approaches to conquer a problem, but incremental improvement will often bring radical ideas back to similar conclusions in the end. A comparison between SPARC and POWER architectures is no different.



Oracle Today: SPARC T4
It seems that Oracle/Sun approached performance from one direction, from massive thread counts & high throughput, eventually growing to fast cores with a critical thread api - so single threaded bottlenecked software gets more hardware resources dynamically (and only when needed or optionally provisioned at the VM layer.) This was available for a year in the SPARC T4, due to pressure from customers for better single thread performance.

IBM Tomorrow: POWER 7+
Then it seems IBM approached performance from the other direction, from massive single thread speed, incrementing cores, and eventually appearing with a physical socket architecture swap of more cores or less cores (and only at purchase time.) Oddly, this option is only being made a year after the SPARC T4 was released, possibly because of pressure from their customers for higher throughput?




Conclusions:
Which way looks better in practice?

That is a good question. When the SPARC T5 is released, around the same time the POWER 7+ is released - the question will be begged... was IBM's new choice offered to the customer at POWER 7+ purchase time better than the 1+ year old choice offered to the customer while SPARC software at run time (or VM restart time)?

We will have to see what the benchmarks suggest.

Wednesday, July 11, 2012

Architecture Update: The ARMs Race


Abstract:
With the explosion of ARM processors in embedded systems, ranging from phones to tablets, ARM designers are creating ever more complex processors, CPU manufacturers are creating more options for system designers, and even system designers are discussing the movement of ARM from embedded to server and desktop systems.

CPU Architecture

64 Bit ARM v8
First, let's discuss the CPU architecture updates to the ARM processing architecture. Traditionally, the ARM processor bundled a 32 bit processor. While this was more than adequate for embedded systems, this limited the application of the architecture into other spheres of computing.



At the ARM TechCon conference in Santa Clara, California, in October 2011 - the ARM v8 64 bit architecture was announced, demonstrating 64 bit extensions similar to what was done with SPARC, AMD, and later Intel CPU's.

The new ARMv8 architecture has two execution states, the AArch32 state that is compatible with prior generations of 32-bit ARM processors, and AArch64, the new 64-bit extensions. At the moment, the ARMv8 architecture has only been profiled for what ARM calls the A line of its Cortex reference designs, which means they are designated for application processing such as that done on smartphones and tablets.
The ARMv8 architecture will bring forward TrustZone virtualization (which debuted with the ARM v6) and NEON SIMD instructions, which debuted with the ARM v7 designs. The interesting thing about the ARMv8 is that it will offer both double-precision floating point math through that NEON unit.
ARM Update: Mali 450 GPU moves from 4 to 8 cores

The Registered published a short article about an ARM roadmap split:
ARM is doubling the punch of its Mali 400 graphics processors with extra cores for tablet, phone and TV makers that are not ready for combined graphics and compute chips.
The microprocessor architect has announced the Mali 450 GPU, featuring eight cores instead of four.
ARM said the 450 showed it remains committed to the 400 range, and said it is now splitting its roadmap.

ARM Designs Quad-Core 32-bit v7
In April of 2011, ARM announced their quad-core Cortex-A15 processor was scheduled to appear in smart phones or tablets in 2012-2013.
ARM's Cortex-A15, however, will up the ante with an out-of-order superscalar pipeline, 40-bit memory-addressing capabilities, floating-point and media-handling improvements, and a clock speed of up to 2.5GHz, all at power requirements said to be comparable to the company's current Cortex-A9 design.
The Cortex-A9 is the design upon which such top-end smartphone and tablet chips as Apple's A5, Nvidia's Tegra 2, and Samsung's Exynos 4210 are based. The Cortex-A15 design, meanwhile, has already been licensed by Texas Instruments and Nvidia, and Nvidia

(ARM v8 Exception Model, courtesy ARMv8 Architecture PDF)
ARM Starts Designing 64-bit
Richard Grisenthwaite, Lead Architect at ARM, provided a technology preview of the ARMv8 64 bit architecture at ARM TechCon in 2011. It was clear from the document that 32 bit ARM processors would continue to be designed and that this was merely a new line of processor design which manufacturers could leverage.

System Designers


(A Boston Viridis server, front view, no cover, courtesy, The Register)

(Boston Limited's first Viridis server, courtsy The Register)



Boston
The Register writes about U.K. IT Supplier Boston is releasing their Viridis platform, based upon their Calexeda partnership, using the Smoothstone ARM processor.
"The Viridis server is using the 1.4GHz variant of the ECX-1000 processors and plunks a 4GB DDR3 memory stick in for each node on the card. The card has two 10GE network ports and four SATA disk ports per processor... The dozen processor cards including memory burn only 300 watts."



(Dell Quad ARM Server Chassis)

(Dell Quad ARM Server Blade)

Dell
It was mentioned during the June 2012 Network Management "System Vendor: CISC, RISC, EPIC Update"  that Dell was breaking into the ARM marketplace. Each Dell blade holds 4 32-bit ARM servers. Ironically, Dell's blade server looks a lot like a far less rugged old Sun Fire B1600 blade chassis, which contained 3U high SPARC RISC SPARC blades, but each Sun blade, from a decade ago, only held single 64 bit server.

(HP Redstone ARM v7 32 bit Server, courtesy The Register)
 HP Enters the ARMs Race
In November 2011, HP released ARM RISC servers to supplement their Proliant CISC servers and Itanium LWIS processors. This was the result of "Project Moonshot".
"To make the Redstone, HP took a half-width, single-height ProLiant tray server and ripped out just about everything but the tray. In goes the passive backplane that the Calxeda EnergyCard, and HP can cram three rows of these ARM boards, with six per row, for a total of 72 server nodes, in a half-width 2U slot... That gives you 288 server nodes in a 4U rack space, or 72 servers per rack unit."
ARM CPU Manufacturers

Samsung
DRAM manufacturer, and more recently cell phone manufacturer, appears to be hiring CPU designers, possibly for the ARM CPU chips, used in Apple and their own cell phones. Various chip designers are being harvested with experience ranging from Sun Microsystems and Oracle to AMD.


Calxeda
November 2011 - Calxeda announced their 32 bit quad-core EnergyCore ARM v7 EnergyCore ECX-1000 Series CPU's.
"Calxeda has spent the past several years tweaking the 32-bit ARMv7 core to come up with its own system-on-chip (SoC) and related interconnect fabric suitable for hyperscale parallel and distributed computing where nodes have only modest memory needs."

[Applied Micro CEP Paramesh Gopi, courtesty The Register]
Applied Micro
Also in October of 2011, Applied Micro announced their X-Gene 64 bit ARM v8 processors.

(X-Gene ARMv8 CPU)
"The X-Gene chip will also include DDR3 main memory controllers, two 10 Gigabit Ethernet ports, SATA storage and PCI-Express peripheral controllers, and a power/management module – all on the same die as the cores."
"The cores will have L1 and L2 caches per core, a shared L3 cache that spans the cores, and have a target clock speed of 3GHz."
"The X-Gene chip also has on-chip CPU and I/O virtualization, just like x86, Sparc, Power, and Itanium chips do. The architecture also allows for various kinds of offload engines to be plugged in and perhaps integrated on the chip package."



(X-Gene ARM v8 Block Diagram, courtesy The Register)
The X-Gene is suposed to be ready to ship second half of 2012 - which is right about now. Taiwan Semiconductor Manufacturing Corporation (TSMC) is first etching the chips using a 40nm process, with subsequent designes in 28nm.


Nvidia
At the Las Vegas Consumer Electronics Show in April 2011, video chip processing giant Nvidea discussed phones based upon their Tegra 2 dual-core ARM Cortex-A9 chips, which bundle graphics processing, licensed the future Cortex-A15 design, and announced "Project Denver" circa 2013 - targeting desktops.
"Denver provides a choice. System builders can now choose a high-performance processor based on a RISC instruction set with modern features such as fixed-width instructions, predication, and a large general register file. These features enable advanced compiler techniques and simplify implementation, ultimately leading to higher performance and a more energy-efficient processor."
Back in September of 2010, Nvidea president and CEP Jen-Hsun Huang also discussed their "Kepler" ARM processor, due in 2011, and the "Maxwell" ARM processor due in 2013.
(Armada XP Processor, courtesy The Register)
Marvell

Chip manufacturer Marvell acquired the ARM RISC CPU business from Xscale in 2006. In 2010, Marvell announced it's quad-core 32 bit ARM v7 Armada XP processor, implemented on a 40nm process.
"...running at 1.6 GHz with a shared 2 MB L2 cache memory... The chip will include variants that support 64-bit DDR2, DDR3, and DDR3 low-voltage memory chips. For on-chip DDR3 controllers, the memory can run at to 800 MHz and ... has ECC memory scrubbing."
"The chip includes four PCI-Express 2.0 x4 interfaces and four Gigabit Ethernet controllers etched into its silicon; it has 16 SERDES lanes for implementing USB, PCI-Express, SATA, SGMII, and QSGMII ports..."
 It seems 2013 could prove very interesting from Marvell.
Conclusions
It is very odd, not to see IBM producing any platforms based upon ARM, but very interesting to see IBM assisting ARM to reduce it's chip process down to 14nm, back in January of 2011. One has to wonder, at what point will IBM stop developing POWER (POWER 7+ is now about 10 months late?) or stop helping ARM produce smaller & faster processors. Up until this point, POWER was not in competition with ARM, but clearly ARM is climbing the food chain, moving to thin client desktops, cell phones, tablets, and now servers.

Apple Mac OSX, based upon BSD UNIX, and Google's Android Linux are the main OS players in the ARM arena - with Microsoft starting to produce Windows ports.

OpenSolaris port to ARM was of interest back in 2009, a code contribution made in 2009, additional work in Feb/Mar 2012 timeframe with some code, Illumos developers considering ARM in March 2012, Google "Summer of Code" ARM project idea suggested in April 2012, a grad student showing interest in April 2012, and with all the activity around ARM servers - one might hope that there will be additoinal interest in the Illumos community.

Will other OS vendors port to ARM?

Wednesday, June 27, 2012

The Processor Market: POWER #1 HPC

The Processor Market: POWER #1 HPC

Abstract:
During June 2012, some very interesting updates happened - some Open Source pieces from Sun and Oracle were combined with the POWER processors to build a new Super Computer. An odd result: IBM's POWER required a lot more sockets to outrun Fujitsu SPARC64... but did so with better power efficiency and using arch-rival Sun Microsystem's (now Oracle's) open source technology.

[wiring 123% more sockets for 55% greater performance, courtesy The Register]
IBM Denies Fujitsu's SPARC64 Year Long #1 HPC Rank!
With a long list of losses, IBM's POWER architecture finally has a win: proprietary IBM POWER architecture now has a #1 HPC Performance spot with Lustre under ZFS - denying Fujitsu their nearly 1 year long spot as the fastest computer in the world, with Fujitsu's fork of Lustre clustered filesystem!

[zfs write performance under linux with lustre, courtesy Lawrence Livermore Laboratory]
Whamcloud, Lustre, and Sequoia Supercomputer

The Lustre clustered/distributed filesystem, formerly owned by Sun Microsystems, now owned by Oracle. It has long been promised to be merged into ZFS. Whamcloud is a commercial enterprise which develops a fork of the Lustre file system. They announced the release of Chroma Enterprise, to bring enterprise management to Lustre.

Whamcloud is using a non-kernel emulated ZFS fork from OpenSolaris. The ZFS implementation still shows linear scalability (in comparison to the native Linux filesystem), as the load increases.

The Sequoia Supercomputer, run by the United States Department of Energy, has an interesting feature - the use of a merged Sun's  ZFS and Sun's Lustre filesystem. Here is a short 30 minute video talking to the PDF from the Lustre User Group (LUG) 2012.

IBM's Tortoise vs Fujitsu's Hare
IBM needed 123% more proprietary POWER CPU sockets to outrun Fujitsu's open SPARCv9 SPARC64 architecture by a mere 55%. The IBM POWER solution proved itself to be about 23% more power efficient, which is truly an achievement, considering how many more sockets were required. The tortoise POWER processor takes less energy than the hare SPARC64 processor.

Fujitsu SPARC64 Loses The Battle of the Alamo
This is somewhat a Pyrrhic victory, kind of like winning the Battle of the Alamo. Could any 1 year old platform hold it's performance position, when the new opposition has a 123% numeric advantage?

This victory was a solid win for IBM, from a supercomputer to supercomputer perspective, but there is an odd conclusion that some people may notice: each SPARC64 old socket appears to demonstrate a minimum of 123% faster than each new POWER socket.

Considering that each SPARC64 socket was an 8 core processor socket, in comparison to the 18 core POWER processor socket (of which 16 cores is usable) - each SPARC64 core is roughly 243% faster than each POWER core!

Fujitsu's SPARC64 Other Battle FrontsThe battles have been continuous since 2011:
SPARC continue to be on the map, in new locations, as well as eating IBM POWER's lunch in smaller installations - for very good reason. The new 16 core SPARC64 chips offer double the performance, in the same socket, making POWER look pale, in comparison.


Better Options for Super Computers
IBM's main processor is POWER with it's main OS being AIX. AIX is lacking a modern file system. IBM had a second operating system option, Linux, but it was lacking a modern file system. IBM briefly toyed with the idea of purchasing Sun Microsystems, before Oracle made the final purchase. AIX and Linux choices on POWER were lacking.

Why was the choice made to emulate ZFS? The licensing in Linux is so restrictive that ZFS could not be combined with the Linux kernel, so it had to be emulated in userland. Why did IBM use Lustre instead of IBM's own GPFS clustered file system? Cost may be a factor and Lustre is basically the defacto standard in High Performance Computing.

Lustre was going to be merged into ZFS by Sun Microsystems, after it's acquisition in 2007. The use of Lustre support directly from Oracle, without hardware, came to an end shortly after the purchase of Sun Microsystems by Oracle. Oracle limited the support of Lustre to Oracle hardware in 2010.

Code changes to OpenSolaris were delivered for Lustre friendliness - the movement to complete Lustre with ZFS under Illumos in kernel space could have offered better performance over user space ZFS, fewer system calls would be required at the emulation layer. Illumos could have delivered native performance on the IBM POWER Sequoia or the Fujitsu SPARC64 K Supercomputer.

Fujitsu, being the SPARC64 creator, was more than capable of delivering their drivers into the Illumos market, had Illumos been interested in SPARC. Clearly, pushing IBM to adopt forks of Oracle's Solaris ZFS and Oracle's Lustre was still pretty aggressive, perhaps pushing them all the way to adopt Illumos, a fork of Solaris, was a bridge too far (especially, after a failed Solaris acquisition.)
Conclusions
With some in the Illumos community seemingly less interested in POSIX subsystems, pulling out SVR4 features, disinterested in non-Intel distributions - some are asking the question the value of Illumos without the differentiators of ZFS and DTrace with an OS like Linux.

With POWER sitting as #1, SPARC64 as #2, and ARM growing with increasing market prevalence - the window for Illumos relevance may be closing if they don't start actively supporting some non-x64 architectures, as their differentiating features get ported to competing OS's.

IBM's POWER has long tried to demonstrate their superiority in per-socket or per-core performance. The POWER platform uses 18 core's per socket while Fujitsu uses 8 cores per socket - so each POWER core is vastly slower than a Fujitsu SPARC64 core.

IBM long tried to demonstrate their superiority of technologies to companies like Sun and Oracle, yet at the core of their super computer was ZFS and Lustre - in order to compete in this arena, former Sun Microsystem (now Oracle) technology was used, to scale their solution.

A non-IBM operating system, running a fork of Oracle Solaris ZFS, and running a fork of Oracle Lustre is not the way some might want to advertize an IBM POWER architecture (which normally runs IBM AIX operating system with IBM GPFS file system.)