Friday, November 30, 2007

FortressITX Selects 3Tera’s AppLogic Grid OS For Utility Computing/Grid Solutions

CLIFTON, NJ--(BUSINESS WIRE)--FORTRESS ITX, the premier boutique Internet infrastructure outsourcing partner for the enterprise, announced today it has launched a new grid hosting platform Dynamic Grid and service plans based on 3Tera's award winning AppLogic Grid Operating System. FortressITX customers can now take advantage of a highly available, highly scalable grid infrastructure to host their online applications.

"With utility computing and virtualization taking hold as a viable IT solution, we see early adopters benefiting the most through shorter time to market and dramatic reduction of deployment and management costs," says Jason Silverglate, President and CEO of FortressITX. Now, by partnering with 3Tera and leveraging their AppLogic grid OS, FortressITX solves all of that with a highly available and scalable way to deploy additional infrastructure in seconds to handle customer's demand for high traffic."

"We are happy to welcome FortressITX to the AppLogic family of partners, because we share the belief that online businesses should be able to easily manage, deploy and scale their applications without being constrained by the complexity and cost of owning and operating their own infrastructure," said Bert Armijo, Senior Vice President, Sales, Marketing and Product Management, 3Tera, Inc. "The new partnership enables FortressITX customers to run Web applications on demand to compete cost-effectively in today's online marketplace."

For more information, visit

Rackable Systems, Inc. Passes $1 Billion Mark in Lifetime Sales

FREMONT, Calif.--(BUSINESS WIRE)--Rackable Systems, Inc. (NASDAQ:RACK), a leading provider of servers and storage products for large-scale data centers, today announced that the corporation has reached the $1 billion mark in lifetime sales of products and services.

Founded in 1999, Rackable Systems has become the fastest growing x86 server provider among the top five providers in North America based on analyst firm Gartners comprehensive review of the market (Servers Quarterly Statistics Worldwide: Database, November 2007), having deployed hundreds of thousands of systems worldwide since its inception.

Hitting the billion dollar mark is a true milestone and a tribute to Rackable Systems leadership in the markets we serve, said Mark J. Barrenechea, president and CEO of Rackable Systems. Our Eco-LogicalTM servers and storage have been deployed for hundreds of customers around the world, and our growth is a testament to the demand for our energy-efficient solutions.

Rackable Systems award-winning server and storage products can help reduce data center operational costs and improve total cost of ownership. With patented innovations in rack density, DC Power distribution and cooling designs, Rackable Systems products help improve efficiency and performance in even the most complex data centers.

Friday, November 16, 2007

First Ever Green500 List Released at SC07

Blue Gene/P Solution and eServer Blue Gene Solution top the list with nine out of first ten. Almost all the Supercomputers that I have access, made the list! Proud to be green and computing!

BLACKSBURG, Va., Nov. 15 -- Virginia Tech released the inaugural Green500 List this morning at the Supercomputing 2007 (SC07) conference in Reno, Nev.

"The Green500 List is intended to serve as a ranking of the most energy-efficient supercomputers in the world and as a complementary view to the Top500 List," said Wu Feng, associate professor in the Departments of Computer Science and Electrical and Computer Engineering at Virginia Tech.

All systems on the Green500 List are ranked by MFLOPS/Watt (million floating-point operations per second per watt). The MFLOPS numerator is the reported LINPACK sustained (Rmax) value recorded by the Top500 List. (LINPACK is a linear algebra software package used to create equations to challenge computers.) The watts denominator is either a direct measurement of the system running the LINPACK benchmark at Rmax load or a peak power estimate based upon machine specifications.

For now, systems must first place in the current Top500 List in order to be considered for the Green500. Of the Top500 machines, more than 200 machines directly reported their measured power for the Green500 List. In cases where measured power was not provided, the Green500 List used peak power, as estimated by the Green500 team, based on the best available specifications for the systems in the Top500 List.

The November 2007 Green500 List is a combined ranking of all 500 machines based on the best (highest) MFLOPS/Watt rating available from either direct measurements or peak power estimations. Because peak power numbers do not necessarily reflect power consumption under load, the Green500 team specifically discourages direct comparisons of measured and peak values in the current Green500.

"As this list is the first attempt of its kind, the rankings are open to interpretation by the media, associated vendors, and the general community," said Kirk Cameron, associate professor of computer science at Virginia Tech. "The Green500 team encourages fair use of the list rankings to promote energy efficiency in high-performance systems. We discourage use of the list to disparage a particular vendor, system, or user," he concluded.

The list itself and the methodology used to rank the systems are works-in-progress, Feng said, adding that this will evolve over time to ensure accuracy and more closely reflect energy efficiency in the fast-paced, ever-changing, high-performance community.

HPCwire Readers and Editors Name Rackable Systems the “Best Price/Performance HPC Hardware Solution” in the Industry

FREMONT, Calif., November 13, 2007 – Rackable Systems, Inc. (NASDAQ: RACK), a leading provider of servers and storage for large-scale data centers, today announced that the company has been awarded a 2007 HPCwire Readers’ Choice Award in the category of Best Price/Performance HPC Hardware Solution. The award honors Rackable Systems’ broad range of high-performing Eco-Logical™ servers and storage, designed to reduce operational expenses and improve TCO in even the most demanding data center environments.

“Rackable Systems is honored to be recognized for delivering compelling x86 solutions to the HPC community,” said Mark J. Barrenechea, president and CEO of Rackable Systems. “Our build-to-order model allows us to tailor each solution to the price/performance needs of each customer, helping to enable the Eco-Logical data center.”

Rackable Systems’ innovative x86 server and storage designs help reduce power and cooling costs while maximizing performance, providing an ideal solution for HPC deployments. With products ranging from Eco-Logical rack-mount servers to the company’s RapidScale clustered storage and its award-winning ICE Cube modular data center, Rackable Systems continues to provide solutions that are customized and built to order to meet the unique needs of any data center. Carefully designed to reduce power consumption and increase reliability, Rackable Systems solutions enable data centers to “go green” without compromising application performance.

“At the University of Florida, we have been pleased with the power and space efficiency provided by our Rackable Systems servers and storage. With their high-density racks and dual-core servers, we were able to house a cluster that would have required three times the space and twice the power of competing solutions,” said Dr. Charles Taylor, senior HPC systems engineer and associate director of the University of Florida HPC Center. “We applaud Rackable Systems for bringing high-performance, environmentally-friendly solutions to HPC environments like our own.”

The coveted HPCwire Readers’ Choice Awards are determined through online polling of the global HPCwire audience, along with a rigorous selection process involving HPCwire editors and industry luminaries. The awards are an annual feature of the publication and constitute prestigious recognition from the HPC community. The 2007 Readers’ Choice awards generated a record number of responses from the several hundred thousand readers who access HPCwire each week.

More information on these awards can be found at the HPCwire website (
Rackable Systems.

SC07, The Super Computing Conference!

I finally managed to sneak in to the SC07, Super Computing Conference that I attend every year, SC06, SC05, etc. I almost missed the opportunity due to prior commitments. But two days are better than nothing and I have a load of information to write about. As usual I learned a lot already there will be more by the end of the day.
SC07, sponsoredby ACM and IEEE Computer Society, will showcase how high-performance computing,networking, storage and analysis lead to advances in research, education andcommerce. This premiere international conference includes technical andeducation programs, workshops, tutorials, an exhibit area, demonstrations andhands-on learning. For moreinformation, please visit

Wednesday, November 14, 2007

SUN Solaris on DELL PowerEdge servers.

"We have 12 million (Solaris) licenses in the marketplace, and a majority of them aren't running on Sun hardware," Jonathan Schwartz, Sun's chief executive, said during a keynote speech at Oracle OpenWorld in San Francisco.

The multi-year distribution agreement was announced live from Oracle OpenWorld, where Sun President and CEO, Jonathan Schwartz and Dell Chairman and CEO, Michael Dell, are keynote speakers. With this announcement, Dell is expanding the range of enterprise-class operating systems it offers to its customers and Sun is expanding the reach of its Solaris OS.

As part of the relationship, Dell and Sun will cooperate on system certification and the development of offerings based on Solaris and Dell solutions. In addition, Dell and Sun have agreed to work together to secure support from key ISVs for Solaris on Dell PowerEdge servers. Both companies will work to ensure the combination of Dell PowerEdge servers and the Solaris OS delivers customer choice and value for applications that demand reliability, security, scalability, performance and integrated virtualization.

"Dell's offering of Solaris redefines the market opportunity for both companies," said Jonathan Schwartz, president and CEO, Sun Microsystems. "The relationship gives Dell broader reach into the global free software community with Solaris and OpenSolaris and gives Sun access to channels and customers across the volume marketplace."

"Part of our focus to simplify IT means delivering customers choice and by adding Solaris to our solutions set we are able to do that," said Michael Dell, chairman and CEO, Dell.

Thursday, November 08, 2007

Bittorrent Distribution System is expanded at ibiblio

ibiblio has added four more open source Osprey bittorrent distribution system. Some of my favorite projects have been mirrored and gone are the speed problems we had on ibiblio. It makes a sense to get what you need without taxing the system a lot. If a lot of people are using the system, the benefits will rise. So grab you project files from ibiblio bittorrent syste.
The ibiblio-hosted or -mirrored projects now serving over now include some of My favorites:

Red Hat Enterprise Linux Available On Demand on Amazon Elastic Compute Cloud

on the heels of Redhat 5.1 announcement, Redhat and Amazon announced a new business deal between the two to provide better product offering to customers.

Raleigh NC - November 7, 2007 - Red Hat (NYSE: RHT), the world’s leading provider of open source solutions, today announced the beta availability of Red Hat Enterprise Linux on Amazon Elastic Compute Cloud (Amazon EC2), a web service that provides resizeable compute capacity in the cloud. This collaboration makes all the capabilities of Red Hat Enterprise Linux, including the Red Hat Network management service, world-class technical support and over 3,400 certified applications, available to customers on Amazon's proven network infrastructure and datacenters.

The combination of Red Hat Enterprise Linux and Amazon EC2 changes the economics of computing by allowing customers to pay only for the infrastructure software services and capacity that they actually use. Red Hat Enterprise Linux on Amazon EC2 enables customers to increase or decrease capacity within minutes, removing the need to over-buy software and hardware capacity as a set of resources to handle periodic spikes in demand.

For more information on the offering, visit

Redhat 5.1 ready for standalone systems, virtualized systems, appliances and web-scale "cloud" computing environments.

Raleigh NC - November 7, 2007 - Red Hat (NYSE: RHT), the world’s leading provider of open source solutions, today announced the availability of Red Hat Enterprise Linux 5.1, with integrated virtualization. This release provides the most compelling platform for customers and software developers ever, with its industry-leading virtualization capabilities complementing Red Hat's newly announced Linux Automation strategy. It offers the industry’s broadest deployment ecosystem, covering standalone systems, virtualized systems, appliances and web-scale "cloud" computing environments.

Red Hat Enterprise Linux 5.1 virtualization delivers considerably broader server support than proprietary virtualization products, and up to twice the performance. This allows greater server consolidation and eliminates a key obstacle to deploying virtualization more widely. And Red Hat Enterprise Linux customers benefit from one of the industry's largest and fastest-growing set of certified applications.

Red Hat Enterprise Linux's deployment flexibility uniquely allows customers to deploy a single platform, virtual or physical, small or large, throughout their enterprise. By providing one platform that spans the broadest range of x86, x86-64, POWER, Itanium and mainframe servers, regardless of size, core count or capacity, customers can gain dramatic operational and cost efficiencies when compared to proprietary solutions. And fully integrated virtualization, included at no additional cost, amplifies these benefits. Notably, Red Hat Enterprise Linux 5.1 provides enhanced support for virtualization of Microsoft Windows guests, providing significant performance improvements for Windows XP, Windows Server 2000, 2003 and Windows 2008 beta guests.

Red Hat works closely with its hardware partners to lead the industry in providing support for new hardware features, an advantage unique to the open source development model. This is reflected in support for features such as Nested Page Tables in the new release.

"Our initial testing indicates that Red Hat Enterprise Linux virtualization delivers significant performance gains for our compute intensive applications and should provide an additional layer of abstraction that will help us manage multiple competing priorities," said Derek Chan, Head of Digital Operations for DreamWorks Animation.

"With Red Hat Enterprise Linux virtualization, customers can easily deploy any application, anywhere at anytime," said Paul Cormier, executive vice president, Worldwide Engineering at Red Hat. "Other virtualization products don't scale to support large numbers of cores or CPUs, which limit customers’ ability to utilize their infrastructure, or force customers to deploy multiple virtualization platforms. With Red Hat Enterprise Linux, customers enjoy a flexible yet consistent application environment for all of their virtualization requirements: from small servers to mainframe-class systems, for Linux and Windows servers and for even the most demanding workloads."

Red Hat Enterprise Linux 5.1 is immediately available to customers via Red Hat Network, Red Hat's management and automation platform. Red Hat Network provides customers a common platform for managing both physical and virtual servers, eliminating the need for organizations to acquire, manage and train their staff on new tools to manage virtual servers. Red Hat Network allows customers to provision, monitor and manage their servers throughout the entire lifecycle.

Red Hat Enterprise Linux virtualization includes the ability to perform live migration, allowing customers to seamlessly move running applications from one server to another, maximizing resource utilization in the face of changing business requirements. Red Hat Enterprise Linux Advanced Platform includes high-availability clustering, storage virtualization and failover software to provide enhanced levels of application availability, for both physical and virtual servers.

Utilizing multiple cores and CPUs is more important than ever with the release of Intel’s latest Quad-Core Intel® Xeon® processors and Itanium® processors. Users deploying Red Hat Enterprise Linux 5.1 and utilizing Intel® Virtualization Technology can experience even greater gains. "Red Hat and Intel have worked together in delivering a high-performance platform for virtualization," said Pat Gelsinger, senior vice president, general manager, Intel Digital Enterprise Group. "Red Hat Enterprise Linux 5.1 allows customers to scale up their virtual infrastructure to run high-performance virtual machines that utilize Intel® Virtualization Technology and all the processing power of the Quad-Core Intel® Xeon® processors and high-end Itanium® servers, without the overhead seen in traditional virtualization environments."

For more information about Red Hat Enterprise Linux 5, visit

Wednesday, November 07, 2007

Record-Setting Fifth Planet Found Orbiting Nearby Star, 55 Cancri, in the constellation Cancer.

55 Cancri, which is 41 light years away and can be seen with the unaided eye or binoculars
The newly discovered planet weighs about 45 times the mass of Earth and may be similar to Saturn in its composition and appearance. The planet is the fourth from 55 Cancri and completes one orbit every 260 days. Its location places the planet in the "habitable zone," a band around the star where the temperature would permit liquid water to pool on solid surfaces. The distance from its star is approximately 116.7 million kilometers (72.5 million miles), slightly closer than Earth to our sun, but it orbits a star that is slightly fainter.

"The gas-giant planets in our solar system all have large moons," said Debra Fischer, an astronomer at San Francisco State University and lead author of a paper that will appear in a future issue of the Astrophysical Journal. "If there is a moon orbiting this new, massive planet, it might have pools of liquid water on a rocky surface."

Fischer and University of California, Berkeley, astronomer Geoff Marcy, plus a team of collaborators discovered this planet after careful observation of 2,000 nearby stars with the Shane telescope at Lick Observatory located on Mt. Hamilton, east of San Jose, Calif., and the W.M. Keck Observatory in Mauna Kea, Hawaii. More than 320 velocity measurements were required to disentangle signals from each of the planets.

"This is the first quintuple-planet system," said Fischer. "This system has a dominant gas giant planet in an orbit similar to our Jupiter. Like the planets orbiting our sun, most of these planets reside in nearly circular orbits."

The first planet orbiting 55 Cancri was discovered by Marcy and Butler in 1996; the planet, the size of Jupiter, was the fourth-known star with an exo-planet ever discovered. It orbits its star every 14.6 days.

The second, third and fourth were discovered over the next eight years.

Read more at NASA JPL

Tuesday, November 06, 2007

Continuous Data Protection for Linux with SteelEye Data Replication for Linux v6

Palo Alto, CA - November 5, 2007 - SteelEye Technology® Inc., a leading provider of award-winning data protection, high availability and disaster recovery products for Linux and Windows, announced today the release of SteelEye Data Replication for Linux v6. The new product is the first on Linux to combine continuous data protection (CDP) features such as "any point in time rewind" with comprehensive data replication capabilities which support multiple replication targets and provide combining data mirrors across any combination of LANs and WANs. The combination of CDP with off-site replication assures businesses that critical data can be restored in case of data corruption, viruses, user errors, or in the face of large-scale physical disasters.

Any Point in Time Rewind
"Any Point in Time Rewind" is a vital CDP element that rewinds data back in time to any moment before data loss occurred. In minutes, data can be restored and business is back to normal. Adding such capabilities to an already robust data replication solution, SteelEye Data Replication also provides organizations with a CDP option that enhances protection of more than one backup of critical business data, whether on-site or remote. Immediately upon noticing data integrity issues, users can mount the replicated data on a backup volume and then move backward and forward through the data stream via time stamps or other user-defined bookmarks until the optimal set of data is constructed. That data are then placed into service and normal operations resume.

Integrated Data Recovery Wizard
The process of managing data recovery is handled through a built-in Data Recovery Wizard. This simple-to-use tool guides users through the building of a new temporary dataset. It validates the dataset for consistency via appropriate tools and then loops back through these steps as needed until the optimal dataset is built. An intelligent binary search technique speeds the rebuilding process to optimize recovery time. Once completed, the wizard places the dataset back into service, onto the production server, and resumes normal business operations.

Multiple Replication Targets
For companies who need the enhanced protection of more than one real-time backup of critical business data, SteelEye Data Replication supports multiple replication targets, which can span any combination of local and remote sites. Local data recovery is assured by continuous backup within the local data center, while asynchronous remote data replication protects against nearby disasters.

More About SteelEye Data Replication for Linux
Leveraging open source features contributed by SteelEye into the 2.6 kernel, SteelEye Data Replication for Linux provides both host-based volume replication and continuous data protection for Red Hat Enterprise Linux and Novell SLES environments. An integrated compression engine along with a block-level change-only architecture reduces network bandwidth requirements and speeds replication.

Used by itself, SteelEye Data Replication delivers real-time backup of business-critical data across either a Wide or Local Area connection. When combined with LifeKeeper for Linux, cluster configurations can be built, ranging from low-cost two node shared nothing clusters to geographically-dispersed clusters for disaster recovery protection.

Monday, November 05, 2007

Dell Pland to Acqiure EqualLogic for $1.4 billion

Round Rock, Texas, and NASHUA, N.H.

Dell has entered into a definitive agreement to acquire EqualLogic, a leading provider of high-performance iSCSI storage area network (SAN) solutions uniquely optimized for virtualization. The acquisition will strengthen Dell’s product and channel leadership in simplifying and virtualizing IT for customers globally. iSCSI SAN technology represents the fastest growing part of the storage business.

“Our customers will be dealing with the largest increase in data we have seen in our history over the next few years,” said Michael Dell, Chairman and CEO, Dell. “Leading the iSCSI revolution will help Dell accelerate IT simplification and virtualization and will drive the Dell value proposition into more areas of the enterprise storage business,” Mr. Dell said.

Under the terms of the agreement, Dell will purchase EqualLogic for approximately $1.4 billion in cash. The acquisition of EqualLogic is expected to close late in the fourth quarter of Dell's fiscal year 2008 or early in the first quarter of fiscal 2009. The company expects the acquisition to be dilutive to earnings per share, excluding the amortization of intangibles, by $0.02 to $0.05 in aggregate for Fiscal 2009 and Fiscal 2010. The acquisition has been approved by the board of directors of each company and is subject to regulatory approvals and customary closing conditions.

After completion of the transaction, Dell plans to grow EqualLogic’s successful channel-partner programs with current and future EqualLogic-branded products, and also plans to incorporate EqualLogic technology into future generations of its Dell PowerVault storage line available through the channel and direct from Dell.

Dell Inc


Thursday, November 01, 2007

massively multi-player Games (MMORPGs) on Grid

Since I wrote about G-Cluster almost two years ago, I have not read too much about Grid and The Games. But a visit to ISGTW made me smile, I want to play Games on my Grid!

Cmputer games have gone hand-in-hand with IT innovation for more than 30 years, capturing the imagination and devotion of millions.

edutain@grid is bringing games to the grid, again.

edutain@grid (Contract No. 034601) is a project funded by the European Commission under the 6th Framework Programme. Its duration is 36 months and started on 1 September 2006.

edutainatgrid: massively multi-player "killer" grid applications