Bovine Powered Servers

As our demand grows for computing power, energy efficiency, and data storage capacity, the ability to produce the power needed in data centers is simply not keeping up with the times. The notion of reducing energy consumption in data centers has led many vendors to pursue increasingly optimistic and sometimes downright quirky ideas. HP is one such vendor which hopes to use sustainable processes in order to build data centers that are self-sufficient. In other words, construct and design a data center whose electricity is generated from a sustainable energy source and whose heat output can be recycled and reused within that same data center. This model aims to give technology companies more options for powering their servers.

In India, for example, electricity is in such high demand that there just isn't enough electricity to keep many of the data centers that are being built there up and running. "In India they need diesel generators because the power grid can't keep up with the growth," said Chandrakant Patel, one of HP Labs researchers. Patel points out that an enterprising farmer with a few cows could be the solution to this energy crisis and even offer a fresh alternative energy approach for IT managers.

So what exactly does a diesel generator have in common with a cow?

As odd as this sounds, cow manure could be a possible solution for small and medium businesses looking for cheaper real estate and electrical alternatives. With the advent of high-speed networks, there is no longer a need to locate data centers within the confines of a big city and can now be located on cheaper land, such as in the rural fields next to a dairy farm. Your average dairy cow will produces 55 kilograms of manure per day which would generate 3 kilowatt-hours of electrical energy. According to HP, a dairy farm with 10,000 cows would produce enough energy to power a 1-megawatt data center, or approximately 1,000 servers.

This recycling process works as follows. Farms already have manure collection systems which utilize anaerobic decomposition methods to breaks down the cow waste much like a sewage treatment plant would. In current systems the biomass goes into an anaerobic digester and after decomposition is released as simple methane gas. However, in HP's vision, instead of a farm burning off the methane gas for energy, which is one of the most volatile greenhouse gases, the chemical energy in that methane could be converted into electrical energy to power the data center. To complete HP's sustainable and self-sufficient vision, the heat given off from data centers will be reused as part of the energy needed to break down the biomass.



This chart provided by HP Labs shows how cow manure and server heat from data centers can be combined to create a sustainable energy alternative.

---

Looking for other Technology Rental information? Visit the Tech Travel Site Map for a variety of computer and technology rental ideas.

Federal Agents Seize $143m in Fake Networking Equipment

Over the past 5 years Federal authorities have seized more than $143m worth of counterfeit Cisco hardware and labels in a coordinated initiative called Operation Network Raider. The operation depends on the collaboration of several law enforcement agencies including the FBI, Immigration and Customs, and Border Protection Agencies which has so far resulted in more than 700 seizures and 30 felony convictions. Despite costing Cisco and other US networking enterprises millions of dollars in sales and technology, the real threat of these counterfeited routers and networking gear is on the level of national security.

In 2008, Ehab Ashoor attempted to traffic 100 gigabit interface converters that were illegally manufactured in China and contained fraudulent documents indicating they were genuinely produced by Cisco. The equipment was destined for the United States Marine Corps and was intended to be used as communication equipment in Iraq. This month Ashoor was sentenced to 51 months in prison and ordered to pay Cisco $119,400 in restitution after being found guilty of trying to sell the counterfeit gear to the US Department of Defense. In January 2010 a Chinese resident was sentenced to 30 months in prison and ordered to pay a restitution of $790,683 for trafficking counterfeit networking gear.

The prospect that our government and business networks may be at risk has propelled law enforcement agencies to work around the clock to crack down on these illegal distribution networks of bogus routers and switches. According to the Customs and Border Patrol there has been a 75 percent decrease in seizures of counterfeit networking hardware at U.S. borders from 2008 to 2009. Yet it is entirely possible that these scams could threaten national security as well as the financial well being of corporations by infusing critical networks with gear that is unreliable, or worse, riddled with backdoors and security vulnerabilities.

China has a well known reputation for doing whatever it takes to get the competitive edge and it has already been proven do–able by researchers at the University of Illinois that such vulnerabilities could be hardwired into a microprocessor. This hacked microprocessor could then log passwords and monitor networking traffic as well as other sensitive data passed through the equipment. However, Cisco has assured us that so far there is absolutely no evidence that such equipment has been tampered with on any scale to contain backdoors, but it is not entirely out of the realm of possibility.

---

Looking for other Technology Rental information? Visit the Tech Travel Site Map for a variety of computer and technology rental ideas.

EPA Delivers Draft 1.0 for Data Center Storage

The US Environmental Protection Agency (EPA) is expected to soon establish the final standards for its Energy Star certification for data centers. Currently the organization is holding sit downs with various storage firms and looking for feedback so that it can move forward with more precise standards for data storage systems such as enterprise hard drives and solid state drives. So far the EPA has already established Energy Star ratings for servers but as you can imagine establishing energy standards for storage solutions is a considerably more complex task. Unlike appliances such as a personal computer or printer, the efficiency of a data storage unit can depend on a wide range of variables such as configuration, controllers in use, power supplies and even software.

The EPA has made steady progress since April of 2009 when it first announced it would be moving forward with the program. Most recently the EPA has been collecting data from December 2009 through March 2010 to gain a better understanding of the dynamic between hardware/software configuration and energy efficiency, active and idle state performance, and sensitivity to single-configuration changes. The EPA has released the results of the research at this stage, entitled Draft 1 Version 1.0 Specification, which can be downloaded for free courtesy of Energy Star. If you're technically inclined the report has some pretty interesting results and may be worth the read.

Draft 1.0 comprises the idea of a "product family" certification, due to the fact that storage devices have a greater level of customization and configurability of products. The report also sharpens several key definitions. For example, the definition of a "storage product" includes components and subsystems that are considered an "integral part" of the storage product architecture, but specifically excludes products that are usually associated with a storage environment at the data center level. Only the storage product can be subject to Energy Star certification -- subsystems and components are not eligible for certification. The Draft also defines Active State, Ready Idle State, and Deep Idle State for those that want to take a look. If you happen to have comments about Draft 1, they're due to the EPA by May 21.

---

Looking for other Technology Rental information? Visit the Tech Travel Site Map for a variety of computer and technology rental ideas.

Former Fujitsu President Sues After Termination

The ex-president of Fujitsu, Kuniaki Nozoe, is now threatening to sue the IT services giant for damages over losses suffered by the company and has even asked the corporation to sue some of its own executives. What prompted the legal action from the former president was his forced resignation last September. This March he wrote the company asking that his resignation be nullified and reversed, a tactic which hasn't gone very well for the former president. In response, Fujitsu alluded that he had been forced to quit due to his ties with organized crime. In fact the board said it had previously warned Nozoe that such links were in conflict with "the Fujitsu Way".

Fujitsu first announced the resignation of Mr Nozoe in September 2009 citing health issues. However last month they admitted that the president had been forced out following an investigation into his business links. The investigation found that Mr Nozoe had a relationship with a third party company said to "have an unfavourable reputation" - a common phrase used in Japan to infer that one has ties to the Yakuza. Nozoe stated that his relationship was merely personal but upon being confronted with the allegations the board and Mr Nozoe agreed with one another to issue a statement attributing his departure to poor health rather than blame the unnamed third party. Although Nozoe did not break any laws Fujitsu maintains that he failed his duties as president.

The episode has since raised questions over the role of organized crime syndicates in big Japanese business. "The suggestion that a major Japanese company has been linked with the yakuza is not surprising," said Dr Seijiro Takeshita, a director at the Japanese bank Mizuho International "Associating with gangsters has often been a part of doing business in Japan - including even the banks." The Tokyo Stock Exchange has since given Fujitsu a strict warning over the issue.

---

Looking for other Technology Rental information? Visit the Tech Travel Site Map for a variety of computer and technology rental ideas.

The Largest Cloud in the World is Dark, Shady, and Criminally Owned

When thinking of the largest cloud computing network known to man, what companies come to mind? Microsoft? Sure they have alot of computers but not even close. Amazon? Getting bigger but still not even in the same ballpark. Google? As monstrous as their cloud is, its a mere drop in the ocean. The largest cloud in the tech world isn't controlled by a brick and mortar corporation, but rather it is a network of computers controlled by the Conficker computer worm across more than 200 countries in the world. So just how big is the worlds biggest cloud?

"Conficker controls 6.4 million computer systems in 230 countries at 230 top level domains globally with more than 18 million CPUs and 28 terabits per second of bandwidth." said Rodney Joffe, senior vice president and senior technologist at the infrastructure services firm Neustar.

In other words the biggest cloud on the planet is controlled by an unknown criminal enterprise that rents out their botnet to send spam, perform a denial-of-service attack, hack computers, spread malware, and steal personal information and money. In fact, it is believed that much of the comment spam that plagues many blogs is spawned from a portion of the conficker cloud. Put simply, the cloud is "mobbed up."

In many ways, the Conficker cloud is much more competitive than legit vendors. The operators have experience with the virus dating back to 1998 and their footprint is bigger than any cloud previously seen. On top of that there are no moral, ethical, or legal constraints with the added bonus of zero costs. There is even an unlimited supply of new resources readily available as the conficker spreads far and wide to take over and steal more computing power.

Just like legitimate cloud vendors, Conficker is available for rent and can be found just about anywhere in the world a user would want their cloud to be based. Users can choose the amount of bandwidth they want, the kind of operating system they want to use, and even what kind of services will be installed into the cloud such as spam distribution, dos attacks, etc.

By the way, just in case you were wondering, the biggest legitimate cloud provider is Google which is made up of approximately 500,000 systems, 1 million CPUs and 1,500 gigabits per second (Gbps) of bandwdith. Coming in second is Amazonwith 160,000 systems, 320,000 CPUs and 400 Gbps of bandwidth. The third largest legit cloud is owned by Rackspace, which offers 65,000 systems, 130,000 CPUs and 300 Gbps.

Although the last major attack performed by the Conficker cloud occurred over a year ago against the Manchester police department, the virus is still considered a very real and palpable threat. If you fear you are infected by the Conficker virus you can try out this Conficker Eye Chart which pulls images from three sites that Conficker is known to block and displays them in a box. If all the images show up you're in good shape, but if one or more doesn't display it could indicate a Conficker or other malware infection. Be aware that if you are browsing from behind a proxy, you may be able to see all the images and still be infected.

---

Looking for a short term file server rental for your next proof of concept or data center move? Call www.rentacomptuer.com at 800-736-8772 today!

x86 Server Market Directs Microsoft to End Itanium Development

Microsoft has announced that it will no longer support development for Intel's Itanium processor effectively placing current Itanium products into maintenance status for the next three years with support ending entirely in eight years. Microsoft also stated that the current versions of Windows Server 2008 R2, SQL Server 2008 R2, and its developer tool Visual Studio 2010 will be the last versions to support the Itanium architecture. For those wondering exactly why Microsoft would make this move, Joe Clabby, President of Clabby Analytics, offers his thoughts on the decision.

"Here's what really happens: Microsoft has invested in x86 architecture. People don't want Windows on Itanium. They want HP-UX on Itanium and maybe some NonStop and OpenVMS, but they have not done jumping jacks over Windows on Itanium. Microsoft is saying its committing heart and soul to x86 multicore and that's what the market wants,"

While the move is yet another blow to the Itanium line, losing Microsoft is not as painful as one would think. Approximately 80 percent of Itanium sales are from HP, which runs HP-UX, NonStop or OpenVMS. Windows and Unix are merely a small portion of their business. However, the marketplace continues to gravitate towards the architecture proposed by Advanced Micro Devices, which added 64-bit extensions to the x86 processors used by many mainstream servers and PCs. Although Microsoft has offered 64-bit versions of Windows Server for both types of chips, the x64 versions have proven to be far more popular than the Itanium ones. Microsoft's reasoning for the decision seems to be sound.

"The natural evolution of the x86 64-bit ('x64') architecture has led to the creation of processors and servers which deliver the scalability and reliability needed for today's 'mission-critical' workloads," Reger said in a blog post. "Just this week, both Intel and AMD have released new high core-count processors, and servers with eight or more x64 processors have now been announced by a full dozen server manufacturers. Such servers contain 64 to 96 processor cores, with more on the horizon."

Despite waning mainstream support and the fact that Itanium has never been a big seller, the chip remains as an importance figure in the market seeing as its the processing power backing HP's high-end server line. In addition, Intel continues to develop new versions of the processor, most recently the Itanium 9300 which was introduced in February, and has promised at least two more generations codenamed "Poulson" and "Kitson". While the immediate future seems secure for the Itanium series of processors it remains to be seen just how far they will be able to go.

---

Looking for a deal on a file server just back from rental? Check out the just back from rental computer inventory at www.rentacomputer.com or call 800-736-8772 today!

AMD - "Welcome to the World of 12 Cores"

AMD has kicked off this week by debuting its new "Magny-Cour" server platform series which includes the new Opteron 6100 8-core and 12-core processors. These are the world’s first 8- and 12-core x86 server processors and come with a host of new features including 4 memory channels, HyperTransport™ technology 3.0, a fourth HyperTransport technology link for better processor-to-processor communication in 4P servers, and new power management features that allow for increased performance when compared to previous generations. The chips themselves began shipping last month but AMD has waited until nearly the end of Quarter 1 to make them official so that Original Equipment Manufacturers(OEMs) would be ready with Opteron 6100-powered machines.

The Opteron 6000 platform targets the 2P and 4P market and is aimed to be used for virtualization, database, and high performance computing applications. Apart from the new CPUs, the Opteron platform features the G34 socket and 5600 Series chipset with I/O virtualization capability, HyperTransport 3.0 and PCI Express 2.0. The Opteron 6100 processors are manufactured on 45nm technology and boast four HyperTransport links, a 4 channel integrated DDR3 memory controller, up to 12MB of L3 cache, and is up to 88% faster than the previous generation of processors.

In today's economic climate, AMD has decided to downplay maximum performance in favor of improved power consumption and a cheaper MSRP. Customers are simply looking for more, not less, in their IT dollar and AMD boasts maximum performance. When placed next to servers of comparable power, competitors fall short in pricing. Based on this graph, consumers are paying 42% more money for the honor of a slower processor. In the power consumption field, AMD clearly shows their efficiency by beating out an Intel 130W Thermal Power Design (TDP) part with their own 80W Average Power Consumpution (ACP) part in terms of performance. Effectively AMD has doubled the cores while staying in the same power and thermal range as previous generations.

In addition to AMD's aggressive 2P pricing, they have upped the value and stripped away the “4P tax.” Long gone are the days when customers are required to pay a premium in order to buy a processor capable of scaling up to 4 CPUs in a single platform. As of today, the 4P "tax" from AMD is effectively $0 but the same cannot be said for other competitors.

"As AMD has done before, we are again redefining the server market based on current customer requirements," said Patrick Patla, vice president and general manager, Server and Embedded Divisions, AMD. "The AMD Opteron 6000 Series platform signals a new era of server value, significantly disrupts today’s server economics and provides the performance-per-watt, value and consistency customers demand for their real-world data center workloads."

The Opteron 6000 platform has already been adopted by HP, Dell, Acer Group, SGI and Cray with many more expected.

---

Looking for other Technology Rental information? Visit the Tech Travel Site Map for a variety of computer and technology rental ideas.