Skip navigation
2008

As processor technology becomes more mobile appropriate, it opens up new doors for all kinds of applications. More processing power makes it easier for these devices to be more intelligent and to communicate wirelessly amongst themselves. The growth in intelligent mobile devices of all types is going to be phenomenal in the coming years. But, many issues exist that hinder the development and growth of mobile devices.

 

To be mobile, they must be small and lightweight. They have unique thermal constraints because of the size. You can't simply throw large heat sinks and heat pipes on them to disperse the heat from the processors and chipsets. Fans are not good. They create a long list of design challenges that are more easily addressed by taking them out of the equation. Who wants a fan making noise and constraining how the device is used so as not to block the airflow?

 

And most visible to users is battery life. Short battery life puts a leash on the mobility factor. Large batteries make the device heavier and less mobile.

 

Improvements in processor and chipset technology are making it easier to overcome these issues.

 

I recently moderated an E-cast event on the topic of "Rethink Cool- Intel® AtomTM Meets Tough Design Requirements" that was sponsored by the Intel® Embedded and Communications Alliance, with presentations from Intel ECG, RadiSys, and Nexcom.

 

Intel ECG started it off with a quick introduction to the Atom processor. RadiSys discussed how COM Express uses the Atom processor to help us re-think cool. The presentation covered the advantages of using a board level module to overcome some of the challenges of designing small systems. The Nexcom presentation touched on techniques on how to design for long battery life in mobile industrial applications.

 

 

You can view the E-cast in its entirety at http://w.on24.com/r.htm?e=117593&s=1&k=D93329A40831838D6947B2A6DCCD85DD

 

 

Several questions were asked during the E-cast. For a list of questions and responses go to Rethink Cool- Intel® Atom™ Meets Tough Design Requirements, Sept 24, 2008, Live Chat

 

 

Message Edited by serenajoy on 03-11-2009 07:56 PM
Message Edited by serenajoy on 03-11-2009 08:01 PM

I mentioned in my last blog that the people behind ATCA have been looking towards the push for next generation technologies. In the case of I/O, this is being driven by new technologies, but the one I will focus on for this entry is power. ATCA is looking to expand into new market areas, while also meeting the increasing capacity demands of its current market space. To do this requires more performance on a simplistic level. And as we know, increased performance tends to require more power, which in turn generates more heat.

 

 

 

Which brings us to “shall” and “shall not.”

 

 

 

One of the most important instances where we reach this crossroads is when considering a change in ATCA specification in terms of power per blade. Initially, ATCA blades were limited to 200W of power, which of course implied that the chassis surrounding it can

cool a 200W blade.

 

However, the latest releases of ATCA-based blades no longer have this restriction. The specification that used to state that an ATCA blade shall be limited to 200W per slot has now changed to a shall not exceed 400W, although at different places within the specification. Obviously, this is an important difference. It allows the ATCA designer to use more powerful CPU solutions to meet that ever increasing curve of capacity requests and enables the ability to support more cores, more memory and more storage. In short, a single “shall” allows ATCA to expand in a much needed direction.

Message Edited by serenajoy on 03-11-2009 08:14 PM

You know what it's like when you have that "A Ha" moment. Well we have been getting a lot of these lately from various customers. What's driving these reactions are the implications of having two low voltage Harpertown processors running on Trenton's MCX/MCG system host boards and what this means to applications that require maximum performance with minimum heat generation. The processors I'm speaking about are the embedded Quad-Core Intel Xeon Processors L5410 and L5408. The performance and thermal design power (TDP) ratings of 50 Watts and 40 Watts respectively have significant implications in the high-end performance segment of the embedded system market that Trenton serves.

 

Now I know that 40W or 50W TDPs sounds horrible compared to the sub-20W TDP ratings common in the low-end, commodity driven portion of the embedded system market. However, the reality we deal with everyday in the embedded system market segments that we serve is a demand for a level of processor performance that in the past has precluded most low power processor solutions. The Intel L5410 and L5408 are rapidly changing this performance vs. heat paradigm.

 

For example, in a surveillance aircraft application, the system needed requires four system host boards with each board having two, Quad-Core Intel Xeon Processors L5408. This system is managing incoming data from a variety of sources, processing all this data and driving the display and communication systems needed to act on these critical inputs. Data processing time and accuracy are critical as well as the need to have a system that generates as little heat as possible. It would have been next to impossible to design a system to meet all of the customer's performance and thermal requirements without the L5410 or L5408 processors.

 

 

Next time I'll share with you some of the system design details regarding the chassis airflow design and the new four-segment PICMG 1.3 Ethernet fabric backplane we produced for this system application.

 

Message Edited by serenajoy on 03-11-2009 08:36 PM
Message Edited by Jim_Renehan on 05-25-2009 06:40 PM
Message Edited by Jim_Renehan on 05-25-2009 06:40 PM

Decisions that Last

Posted by Kontron_Pantone Sep 22, 2008

My best friend sold her house today. In the midst of the

gloom and doom of foreclosures a normal house sale went through. It got me

thinking about the high tech business where there is still plenty of buying and

selling going on if the products fit the needs. In today's environment,

companies making major investments in embedded industrial servers

want to get the exact configuration that suits their needs.

Just as my friend selected the flooring and cabinets in her new home,

companies should select a vendor that offers a full range of CPU/chipsets,

memory, drives and accessories for embedded multicore server systems purchases.

 

 

 

 

 

 

 

Moving single core applications to multicore processing

involves investing resources to take advantage of the resource allocation and

capacity management capabilities of multicore. It is a lasting decision. The

payoff is improved application performance. Selecting the right system

configuration is part of the transition process.

 

 

 

 

Once the decision to migrate an application to a multi-core

industrial server has been made, that decision should not have to be re-evaluated just

to deploy additional industrial servers. Kontron realizes the importance

of helping customers choose the right server for their long term needs

and the value of long term support.

 

 

 

 

 

 

Kontron - Nancy Pantone

 

 

Message Edited by serenajoy on 03-11-2009 08:37 PM

Did you know that the RadiSys Procelerant™ CEGM45 COM Express family supports both the Intel® GM45 and Intel GS45 chipsets? The chipsets have the same functionality, but the GM45 was released in July with the larger 35mm x 35mm package while the GS45 was released August in the smaller 22mm x 22mm package. The GM45 is validated with the high performance 2.53GHz Intel® Core™ 2 Duo T9400 processor while the GS45 is validated with the Intel® 1.86GHz SL9400, 1.2GHz SU9300, 2.26GHz SP9300 Core™ 2 Duo and Intel® Celeron® 722 processors.

 

Why does this matter? It matters if you need the performance of the T9400 processor. The GM45 and GS45 require different board designs due to the package size. The RadiSys Procelerant CEGM45 family contains two board designs so you can have the option of choosing the processor that best meets your price / performance / power needs. From a user perspective - I don't care if it is a small or large package – I just want the processor module that optimizes my system.

 

We couldn't ignore the T9400 processor. The combination of the T9400 processor and GM45 chipset hit the higher end processing needs of processing-intensive applications such as medical and machine imaging, test and measurement, and communications platforms. Feature such as 8GB DDR3 memory, 6MB cache, and SSE4 SIMD instruction enhancements, four SATA ports, and Gigabit Ethernet combine with the T9400 processor to maximize the performance vector of the Procelerant CEGM45 product family.

 

So, small package, large package, what ever package – know that RadiSys has already taken care of it for you. Your processor options are wide open.

 

Jennifer Zickel

Product Manager, COM Express Product Line

http://www.radisys.com


Message Edited by serenajoy on 03-11-2009 07:57 PM
Message Edited by serenajoy on 03-11-2009 08:01 PM

It wasn't that long ago that all eyes were on processor clock speeds. It seemed like every week a faster processor was announced. First it was a mad scramble to 1GHz. Next, the race was on for 2GHz. When could we expect 10GHz? It was mesmerizing. Sadly, even though raw Megahertz is an unreliable predictor of performance, many design decisions were driven almost entirely by these numbers. Fortunately we've learned a bit from this, right?

 

 

We've now entered an era of focus on number of cores. First it was dual-core, then quad. Who will be the first company to offer one-hundred cores? When can we expect a thousand-core processor? Suddenly it's all about cores. Again, I feel the familiar draw to adopt a "more is better" approach to product comparison. Perhaps you do too. Let me offer the following tips to help resist this temptation.

 

 

Not all Processors are Created Equal

 

 

First and foremost, processors are designed with a purpose in mind: DSPs are designed to perform signal and image processing tasks, packet processors are best suited for manipulating network headers and routing traffic, etc. Processor specialization comes from unique instruction sets, architectural tuning, and hardware acceleration engines. The bottom line is that processors will perform best on the applications for which they were designed, and might perform quite poorly in other uses.

 

 

This isn't to say that a packet processor or communications processor can't run your server-class application; however, this should be approached with caution. You probably have an uphill battle ahead of you regardless of the number of cores available. Processor type, not number of cores, should be the first criteria in selecting a solution.

 

 

You Can't Get Something for Nothing

 

 

Secondly, processor designers have a limited number of transistors they can use (determined by semiconductor process) and more cores come at the expense of something else. A fifty-core processor designed using the same semiconductor technology as a four-core core processor must have gotten rid of something. In many cases, the processor cache is the first thing to go. After that, floating point, machine word width, and instruction set depth are all candidates for the chopping block. In order to make a reasonable comparison, you need to determine how important each of these features is to your design.

 

 

Selecting the right multicore processor for your application doesn't need to be a complicated process. Guided by these tips you can probably avoid the pitfalls associated with simply counting cores and hoping for the best. If however you would like to delve into this topic in more depth, I invite you to post to this blog or contact me directly.

At recent conferences and exhibitions we (Emerson Network Power) have shown a video clip we made to demonstrate the idea that MicroTCA is suitable technology as a base for various rugged applications. Ahead of the official Rugged MicroTCA specifications, this video was created to show what might be possible in an air cooled environment.

 

Previously we’d developed a video clip showing a proof of concept MicroTCA system using totally standard AMCs operating while being vigorously shaken on a vibration table in our lab. This was pretty impressive even though the vibration was limited to a single axis. This time we wanted to widen the scope to show a more realistic scenario that included temperature. Since most automotive companies test their prototype vehicles in the desert near Emerson’s Embedded Computing headquarters in Arizona, where summer temperatures often exceed 110ºF (whew!), we used the same environment to demonstrate the system’s ability to withstand extreme temperatures.

 

We used an M561 Gama Goat vehicle for our somewhat unscientific, but nevertheless interesting test. The Gama Goat is a six-wheel drive, semi- articulated vehicle that was widely used by the US military from the 1960s to 1980s. It has exceptional off-road manoeuvrability while generating incredible amounts of noise and fumes from its rather environmentally unfriendly diesel engine!

 

We mounted one of our standard PrAMC7210 AMCs, based on the Intel Core 2 Duo processor and a physical AMC hard disk inside a ruggedized MicroTCA enclosure from Hybricon. Running a standard OS, the processor AMC was set up to capture video from an external webcam connected via a USB port. The ruggedized Air Transport Rack (ATR) was physically bolted to the roof of the Gama Goat using some metal brackets along with the webcam, which was positioned to show the ATR box in the foreground with the environment in the background.

 

We hit a temporary problem with applying power to the system; due to time limitations and availability of connectors we couldn’t make use of the rugged connectors on the front of the ATR box. We had to run cables directly into the top of the box, which meant we had to leave the lid off, mitigating any cooling effect of the forced air fans and instead exposing the AMC boards to the full effect of the sun.

 

I’m not sure who was the hottest: the driver of the Gama Goat, the camera team who had to climb the desert hills with their equipment or the AMC modules. Either way, they all survived and as you can see from the video clip, the Intel AMC board recorded some great footage. Since we’ve now recovered, we’re looking for the next idea to show off a MicroTCA system………..ideas on a postcard please?

 

A snippet of the full video in WMV format can be downloaded from the resources section at http://communities.intel.com/docs/DOC-1990

We have been designing single board computers for compute-intensive industrial applications at Trenton for long time now. Is there anybody out there old enough to remember the 8088 processor? I wouldn't say that "we've seen it all" over these past thirty years, but we've seen our fair share of processor advances and architecture changes. The one thing that has remained constant over this time period is that as processing speed and capability kept increasing, so did the heat generated by the CPU. That is, until now.

 

We're very pleased that this processing capability vs. thermals SBC design paradigm at our end of the embedded computing application spectrum is finally starting to shift in the customer's favor with the advent of multi-core processing processor architectures from Intel®. In our product line the Trenton SLT/SLIseries of PICMG® 1.3 single board computers or system host boards (SHBs) were the first products to use low power, dual-core Intel® Xeon® processors to provide superior system performance with a low-profile thermal solution. The low-profile heat sinks and the lower CPU power ratings of these SBCs has enabled us to deliver system solutions that use a six-segment backplane to incorporate up to 24 processing cores in a single 4U, 19" rackmount computer. Here are some of the system design details on this Cluster Computing.

 

 

 

 

Next time we'll discuss the systems we are delivering with Trenton MCX/MCG series SBCs using the Quad-Core Intel Xeon Processors Series 5400 (a.k.a. Harpertown) in a few military and  Internet protocol (IPTV) applications.

 

 

Jim Renehan

Trenton Technology Inc.

 

 


 

 

Message Edited by serenajoy on 03-11-2009 08:38 PM
Message Edited by Jim_Renehan on 05-22-2009 09:06 AM
Message Edited by Jim_Renehan on 05-22-2009 09:07 AM
Message Edited by Jim_Renehan on 05-22-2009 09:10 AM

Network threats require enhanced security for access control, user authentication, and attack protection, concerns which require a leap in performance-particularly VPN performance. VPN performance is critical, yet many medium-sized businesses have typically been priced out of VPN acceleration, resulting in compromised features and performance.

 

To address this, a new breed of platforms based on Intel's new EP80579 Integrated Processor deliver untouchable performance for less than half the price of previous platforms. With as much as 1600 Mbps of VPN throughput, they deliver a "no-compromises" approach to security for medium-sized businesses.

 

Advantech's new Intel® EP80579-based FWA-3240 illustrates these advantages. Initial results yield 1600 Mbps IPsec VPN throughput, with as little as 10\% CPU utilization, power reduction of almost 20\%, and decreases in board size of nearly 45 percent.

 

OEM's can forgo specialized co-processors and dedicated security hardware while remaining cost-effective (up to 50\% reduction) and extremely power-efficient.

 

The SoC is backwards code-compatible with earlier Intel® processors allowing security vendors to run existing software applications on Intel® EP80579.

 

 

Intel® EP80579 delivers performance without sacrificing programmability, providing enough CPU margin to respond to dynamic threats while offering the capacity for additional value-added software services. Medium-sized businesses can benefit from VPN acceleration without having to compromise on features and performance.

 

 

Compared to past solutions, Intel® EP80579 offers dramatic improvements in cost, power, and board space, while offering major advances in throughput and headroom. With all of these advantages, Intel® EP80579 is set to revolutionize the network appliance market.

Embedded boards are shrinking at a fast rate. Shrinking in size, that is, and growing in volume. More performance needs to be packed into a smaller space to support the portability and size of the end equipment that these boards are designed into.

 

Form Factor

PCB Size(mm)

Micro-ATX

171 x 171

Mini-ITX

170170

EPIC

165 x 115

Nano-ITX

120x120

Pico-ITX

100 x 72

COM Express

95 x 125

PC104-Plus

96 x 90

Mobile-ITX

75 x 45

 

The trend towards small also lends itself towards custom. Cost is king for handheld or portable equipment - and one way to cut cost is to cut features that are not required. There are two paths; build ground-up as a custom, or start with a standard and reduce. Ground-up custom probably will reach the lowest cost but have by far the highest development cost, paid for by the customer. Alternatively, the design can start with a standard product and for a reasonable fee, de-populate the unnecessary components and connectors to get to an optimal high volume cost point.

 

RadiSys designs small form factor boards with depopulation options for high volume, cost conscious opportunities. Starting with a standard is the fastest option to get products to market the fastest. Saving the depopulation option for when the product reaches high volume production, saves cost and time. Options are good.

 

Peter Mitchell

Sr. Product Line Manager


Message Edited by serenajoy on 03-11-2009 07:59 PM
Message Edited by serenajoy on 03-11-2009 08:02 PM

… and there was a parallel bus, which was fast and effective. But as time passed and speed increased, the poor old parallel bus could not keep up and next generation serial solutions started to appear. At the same time, new processors with increased capability, capacity and performance arrived, but required much more power than before. Supplying the power and cooling for these new technologies was becoming an issue.

 

As people looked forward, they saw that the current solutions were inadequate to support the new performance requirements. Additionally, the advent of the packet based network, deprecation of custom designs and the dawn of COTS infrastructure were all considered as they designed a new specification to meet those new requirements. And so, ATCA was born.

 

Designed for the next generation packet based networks with NEBS compliance as its basis, ATCA is a serial based solution with large blades that can provide the power and cooling support the new CPUs needed. The base specification covers power distribution (aligned with central office needs), management structure, managed field replaceable units, interconnect performance, flexibility, expansion and redundancy to name a few.

 

Now, at the time, there were many possibilities of how these blades could be interconnected. With Infiniband, PCIe/ASI, sRIO, Hypertransport and (of course) Ethernet as possibilities, there were a number of strong candidates. Therefore, the ATCA specification was structured to allow these to be applied as dot specs, additional overlays on the base specification to provide the required customisations for handling the protocol. In hindsight, it would have been easier to pick one, of course, but at the time it was not an easy choice. Today, it is easy to see that Ethernet has become the king of backplane interconnects for ATCA and has driven a strong and flourishing ATCA eco-system. The other interconnect standards are now being used more for onboard interconnect and specialised applications.

 

With ATCA now firmly established as a key solution for telecommunication equipment manufacturers – some of whom even base their standard platform strategy on the open standard – it is imperative that we focus our attention on the future of ATCA. In coming blogs, I will look at several aspects of the push toward new technologies and the next generation of ATCA.

Filter Blog

By date: By tag: