Core i7-975 Extreme Details

The Core i7 vital stats you already know apply here, just as they have since the architecture launched last year. Manufactured on Intel’s now-mature 45nm process, a single Core i7 die populates 263 square millimeters.

Natively quad-core in that the processor doesn’t consist of two dual-core die on a single multi-chip module (like Core 2 Quad), Core i7 also includes Hyper-Threading technology. The result is a micro-architecture with four physical execution cores able to concurrently work on eight threads. As a result of the efforts expended by software developers to better-optimize relevant code for parallelism, this results in a performance win more often now than it did back when Hyper-Threading first emerged in the Pentium 4 days.
Cache sizes remain the same (32 KB L1 I/32 KB L1 D and 256 KB L2 per core, plus 8 MB shared L3), and the integrated triple-channel memory controller is still officially limited to DDR3-1066. But of course, as we’ve discovered, retail CPUs support the multipliers necessary to reach as far as DDR3-2133. We have heard from one memory vendor that the controller itself has been improved, but without any additional information from Intel regarding how it might have been tweaked, we can’t confirm those rumors at this time. We can say that DDR3-2133 is now within reach, though it takes some serious tuning to stabilize at that data rate.

As with the i7-965 Extreme, the 975 boasts a 6.4 GT/s QPI link, while the i7-950 employs a 4.8 GT/s link. Of course, if you’re running a retail processor (and not an engineering sample, like the one used in our original Core i7 launch coverage), you should be able to manually tune QPI speed up to 6.4 GT/s in your motherboard’s BIOS.

We asked Intel about the i7-975's Turbo bin configuration and were told that it is exactly the same as the i7-965 before it. That is to say, when 1, 2, 3, or 4 cores are active, you get 2, 1, 1, and 1 available bin (a bin being 133 MHz). Curious as to how much time our 975 Extreme would spend at 3.6 GHz, we ran a single thread of Prime95 to tax an individual core. Interestingly enough, you spend a lot of time waiting for that 27x multiplier to kick in (up from 25x), and it doesn't last very long. You see, there's always something else going on in the background, and if there isn't a significant load being applied to at least one thread, SpeedStep is throttling you back the other way. Expect most of your load time to be spent at 3.46 GHz with Turbo mode enabled. Otherwise, turn the feature off completely and overclock manually.
As a result, we have to wonder how much benefit upcoming architectures will see from Turbo with a single core active.

Hard Drives, Yesterday And Today: From 500 GB To 1.5 TB

Hard drives have now reached the 2,000 GB (2 TB) capacity level, and performance has been steadily going up as well. Hard drive makers have finally incorporated power consumption into their design decisions, making modern hard drives not only bigger and faster, but also more efficient when looked at either from a performance per watt or capacity per watt standpoint.
We took the last three Samsung desktop hard drive product generations and compared their top models to analyze how much progress has been made in the hard drive space.

Hard Drives Versus Solid State Drives
The most recent solid state drives, which are referred to as flash SSDs, have reached capacities of up to 256 GB, and their performance often exceeds 200 MB/s with extremely short latencies. However, only a few of them are truly worth the several hundred dollar investment they demand, as flash SSDs require intelligent, multi-channel configurations with smart controllers and cache memory. The cache is required to enable command queuing, in an effort to maximize wear leveling and performance with changing performance loads. But we’ll stop talking about flash storage, as it is only interesting in the very high-end and the very low-end. Hard drives will continue to dominate the storage market for several years.
Capacities of up to 2 TB cannot yet be realized on flash memory; and if it were possible, it would cost thousands. The cost advantage in the mainstream is even more significant, as terabyte hard drives are available at only $100, while you have to spend three times as much for only 10-20% of the capacity on flash SSDs. And finally, the flash market could not even supply sufficient flash memory to saturate the storage demands of today (and tomorrow).

Samsung: From 0-60 Within A Few Product Generations
Most people don’t think about Samsung when they talk about hard drives, but the Korean company has managed to become an important player, next to Seagate, Hitachi, and Western Digital. The Japanese companies Fujitsu and Toshiba are still pretty active, but they mainly focus on notebook drives (Fujitsu, Toshiba) or server hard drives (Fujitsu). The latter also applies to Seagate and Hitachi. Samsung and WD have server offerings, but their product lines are limited.

Desktop Hard Drive Analysis
We will look at some notebook hard drives in a future article, as these HDD types will dominate the storage market in coming years, due to the shift from stationary to mobile computing. Today we’ll look at three hard drive generations by Samsung: the Spinpoint T166 at 500 GB, the Spinpoint F1 EcoGreen 1000 GB, and the Spinpoint F2 EcoGreen 1500 GB. These represent Samsung’s last three product lines, and they serve as perfect examples to pinpoint where storage is heading.

SKorea military networks under growing cyber attack

South Korea's military computer networks are under ever-growing cyber attack with 95,000 cases reported daily on average, officials said Tuesday.

The Defence Security Command said in a report to a security forum that every day the military counters an average of 10,450 hacking attempts and 81,700 computer virus infections in addition to other cases.
The attacks increased 20 percent this year compared to 2008, it said.
A spokesman for the command told AFP most of the attacks are the same as ordinary people experience at home, but one-tenth are serious.
"Eleven percent of the total are sophisticated and vicious attempts to hack into military servers and to gather intelligence," the spokesman said.
The command did not elaborate where the cyber attacks originated. Defence officials in Seoul have previously pointed to North Korea and China, which they say run elite hacker units.
Yoo Ho-Jin, an official of the National Intelligence Service, said his agency recently proposed that the president name an aide to deal with cyber-security.
"Our country continues to be vulnerable. Some of our government branches failed to function when we recently simulated a cyber-attack on them," Yoo told a security forum on Tuesday, according to Yonhap news agency.
"This is a grave threat to our national security."
South Korea and the United States in April agreed to cooperate to defend their defence networks from countries including China and North Korea.
Last year South Korean Prime Minister Han Seung-Soo warned his cabinet against what he called attempts by Chinese and North Korean computer hackers to obtain state secrets.

The WAN Traffic Controller Juggling Act

Boosting the speed of wide area network application delivery by directing traffic through multiple network paths is a little like running an airport with multiple runways. The more runways, the less congestion. However, both scenarios require someone to direct traffic. Efficient WAN directing devices put you in the control tower.

The Internet has become a critical component in today's fast-moving business environment and continues to play a central role in delivering mission-critical business applications and vital communications to employees, partners and customers. The Internet holds the promise for organizations to streamline operations, improve operational efficiencies and lower costs. As a result, businesses have come to rely on networked applications delivered over the Internet for day-to-day operations and as a means for gaining competitive advantage.
Because of the many possibilities for improving business operations, IT personnel are placing increased attention on application service delivery and the increasing role that their organization's WAN (wide area network) plays within the application delivery ecosystem. The WAN is a critical component of today's business infrastructure. However, the WAN is independent of the business, being under the control of the telco and Internet service provider (ISP), which is an important issue that all businesses that rely on the Internet need to address.
The high value that WANs possess is a direct result of the consolidation of the data center, the centralization of user applications, increasing mobility of employees, the need for business continuity, and the addition of IP (Internet protocol)-enabled applications such as VoIP (voice over Internet protocol) and CRM (customer relationship management).

WAN Link Controllers Give Businesses Control Over the WAN
By routing traffic over multiple service provider links, WAN link controllers create a redundant WAN architecture that provides reliable network uptime while directly impacting the performance of applications over the Internet. This improvement is enhanced by capabilities such as link load balancing and failover and the ability to shape bandwidth for specific applications. Integrated firewall, VPN (virtual private network) and denial of service (DoS) security also help to ensure that the associated reliability and performance gains do not get thwarted by security attacks.
WAN scalability and cost reductions are realized by the ability to have complete freedom of choice for ISP/telco connectivity, allowing network designers and administrators to deploy a variety of cost-efficient bandwidth options such as T1, cable, wireless, DSL (digital subscriber line), etc. The simplicity of adding and removing WAN links and service providers and the efficient use of existing connectivity resources through link load-balancing and bandwidth management techniques makes WAN link controllers so valuable. Additionally, these products are far more cost-effective and simple to deploy than trying to use Border Gateway Protocol (BGP) for bi-directional link load balancing.
Multimedia and other bandwidth-hungry applications that performed well over a local area network (1 Gbps) with ample capacity are being challenged to provide the same level of WAN performance over a T1 link (1.45 Mbps) or DSL link (500 Kbps). This is especially true when the WAN is bottlenecked, and as more applications such as streaming video and VoIP get deployed over the same WAN link. SMEs (small to medium enterprises) need a way to optimize bandwidth in order to meet traffic requirements and improve traffic flows.
Today, the majority of articles written about WAN optimization tend to focus on technologies such as caching, compression and protocol acceleration. However, even with all the compression, caching and protocol acceleration money can buy, if the WAN link that the applications are running over fails, the applications will not get delivered -- period. In today's dynamic business environment, IT personnel need to have a network infrastructure that is redundant, flexible, scalable and can apply appropriate levels of bandwidth to specific applications running over the WAN.
Boosting the speed of WAN application delivery by directing traffic through multiple network paths is surprisingly straightforward. Using an airport analogy, adding a second runway can significantly improve traffic flow from such causes as airplane breakdowns, congestion and poor weather conditions. Similarly, the effectiveness of the multiple-path approach for WAN connectivity depends greatly on the traffic directing devices (i.e. WAN link controllers) and their ability to efficiently and accurately detect network congestion, make routing decisions to circumvent bottlenecks and prioritize applications with bandwidth guarantees. This needs to be accomplished in a manner that is relevant to the specific types of traffic payload being routed. For example, with VoIP traffic, a high-latency network path should be avoided; or for large file transfers, a low-bandwidth path should be avoided.

QoS Prioritizes WAN Connection Bandwidth for Mission-Critical Applications
Many WAN link controllers include the ability to manage applications over the WAN using traffic shaping and quality of service rules, allowing administrators to define traffic and application limits and enable application queuing to prioritize different traffic types. This allows greater control of available bandwidth so that high-priority applications such as VoIP are allocated the bandwidth they require for optimal delivery over the WAN. Bandwidth usage can be managed based upon business policies that are associated with specific mission-critical applications in order to avoid bandwidth contention.
In the first example here, an Internet-based business has two ISP connections using BGP for high availability. Its primary applications are VoIP and email. The applications run smoothly when the primary ISP connection is available. However, when that connection fails, the second ISP connection handles all the traffic, and the VoIP application utilizes most l of the available bandwidth, depriving the email application of bandwidth. This causes a business disruption, as a significant amount of their business is conducted via email, which dramatically impacts productivity and business communication.

In the second example below, the WAN link controller uses QoS to prioritize the VoIP and email applications when the ISP connection is restored. By dedicating bandwidth for each application, the WAN link controller ensures that the applications will have bandwidth allotted to them, enabling each to have the bandwidth they need for optimal service delivery.

WAN Link Controllers Put You in Control
By moving intelligent switching functionality to the edge of an enterprise network, WAN link controllers provide administrators with a new level of control. Administrators can dynamically direct traffic based on service provider availability, line capacity, performance and other policies.
WAN link controllers are deployed in-line between gateway routers and firewalls. To monitor WAN connectivity status, they perform transparent health and performance checks to evaluate the quality and reliability for each ISP link. Using this information, the WAN link controller intercepts traffic flowing in and out of the LAN and automatically switches users to the optimal WAN links.
Where traffic is sent is determined via advanced algorithms that take into account elements such as bandwidth utilization and other criteria, including what an organization pays for each ISP link. Administrators can set these policies to define how traffic should be directed to service provider links in order to best leverage its bandwidth investments. The WAN link controller prioritizes traffic to achieve optimal application delivery. For example, the IT department can make decisions whether to prevent (or limit) the download of YouTube content and other non-business related Internet browsing on the WAN network during business hours. Or they can set up higher bandwidth priorities for mission-critical applications such as VoIP. As applications such as VoIP continue to grow, the need to optimize and efficiently manage them over WAN networks becomes critical. To that end, the WAN link controller allows administrators to optimize VoIP transmission by routing VoIP application usage based on IP address, traffic type, user address, etc. For example, all the VoIP traffic can be allocated on a single line and have a second line allocated to aggregate the combined bandwidth in order to accommodate the traffic load.

Microsoft at E3: Look, Nintendo, No Controllers!

Microsoft wowed the E3 video game convention with a demonstration of Project Natal, a motion-sensing Xbox technology in development. With Nintendo's popular Wii, players use a handheld wand; however, Natal uses cameras to sense a player's full body movements. Despite the riveting demo, Natal is still unproven in terms of its actual gaming performance, and release timing is up in the air.

Microsoft is controlling the early buzz at the massive E3 video game convention with its Project Natal technology, which allows gamers to interact with their Xbox 360s without the need for handheld controllers.
Along with the publicity, though, the company is creating a lot of questions regarding its ability to deliver on the innovation and promise demonstrated on a Los Angeles Convention Center stage Monday.

It made sense for Microsoft to invite Steven Spielberg to help introduce Natal to the throng of media and industry insiders in the audience. Natal uses a full-body motion capture camera along with voice and facial recognition technology to let users do their best impressions of Tom Cruise's character from Spielberg's 2002 science fiction opus "Minority Report." Like Cruise, Natal players will use their movements and their voices to interact with games and, as Spielberg hinted, possibly other forms of entertainment.
Natal demonstration videos now on the Web feature a family member pretending to hold a race car's steering wheel from her living room couch while zooming along a Formula One course in a game; another gets martial arts lessons from an in-game teacher who follows a teenage boy's movements as he paces the floor; that boy's little brother later controls a Godzilla-like monster on the screen. The parents are seen using hand gestures to pick their movie choices from an Xbox Live catalog.

The Big Questions
All of this is the stuff that a gamer and a first adopter's dreams are made of, naturally. So when can you buy it and how much will it cost? Will the games featured in the demonstration be available, and so on?
Microsoft had no answers to any of these questions, but there are plenty of unofficial efforts to fill in the blanks. Microsoft executives did say that the software development kit for Natal is now available for third-party game developers.
Given the long lead time required to bring games to market, it's unlikely that Redmond will be giving a Christmas present to the gaming industry this year, said Directions on Microsoft vice president Matt Rosoff. A pre-holiday release "could be interesting and could spur game sales and keep console sales strong," Rosoff told TechNewsWorld, "but if it waits another year, we'll have to see if the buzz dies down."
If Natal were to come out before the end of 2009, the games available would all have to be from Microsoft studios, Rosoff speculated, as third-party developers would need more time to build games based on the new technology -- and Microsoft would want a lot of consumer choices available at launch.

Demo Is One Thing, but What About the Living Room?

Natal looked great in the E3 demonstration and on the video presentations, Rosoff acknowledged, "but of course, we're going to have to see how it works in the real-world environment. They are camera-based sensors, so light and the amount of light in your room might be an issue. It's just one of those products that until you start to see some hands-on reviews, it's going to be hard to know whether it works as advertised."
Natal is a good answer to the Wii-mote, which has helped Nintendo vault back to the top of the video game industry, said Rosoff. However, Xbox has come a long way from the 2001 introduction of its original console, and now its 360 version has reached a solid No. 2 position in terms of sales, ahead of former leader Sony (NYSE: SNE) and its PlayStation 3.
"I think all three companies want to keep this generation of consoles around for a while, so there should be no new consoles or arms race. The competition will be in services and peripherals," Rosoff predicted.