feature article
Subscribe Now

Help! Help!

Xilinx Premieres Premier

In the years we’ve been covering FPGAs, the technology and the market have been expanding in all dimensions. The devices themselves have grown exponentially bigger, faster, more capable, and more complex. The number and variety of applications have expanded too – with new application areas annexed into the FPGA arena on a regular basis. FPGAs have branched out from simple glue logic to complex system-on-chip integration devices in a wide gamut of markets and systems. 

The tools for FPGA design have followed suit, with incredible pressure for tool developers to provide both increased automation and ease-of-use at the same time they were adapting to devices with orders of magnitude more complexity. Tool suppliers had to take FPGA design tools that could manage 20K LUT glue-logic designs in the hands of an expert and evolve them into tools that could assist a novice in developing a complex FPGA-based system-on-chip with FPGAs ranging over a million LUTs. This is not an easy task.

If you thought that keeping up with such vast changes sounded like a daunting challenge for FPGA companies, you’d be right. FPGA vendors have actually done a respectable job tracking this wild race and keeping their expanding base of customers – old and new – succeeding with each successive product generation. However, FPGA companies and their customers could never hope to go it alone. For many system designers, their needs in the FPGA space far exceed their team’s FPGA design capabilities. These teams don’t have the luxury of spending a couple of years honing their FPGA skills before embarking on their product design. Their best option is to engage consultants that are already FPGA experts.

But where to turn?

There are plenty of consultants and consulting firms hanging out shingles that essentially say, “FPGA designs – Done Here – While You Wait.” Of course, the key to success or failure in your project may lie in choosing the right consultant with the right expertise – not only in FPGA design, but in your application domain as well. Furthermore, you want a consultant that has experience with the tools and technology of the particular FPGA you plan to use. When you’re paying for an expert, you don’t want to have to pay to educate them as well. 

Design services is big business. How big? Today, outsourcing electronic design is estimated at over one trillion dollars, and it is expected to grow to 1.4 trillion by 2020. That’s a lot of shingles, and a lot of consultants to wade through, to find the perfect match for your FPGA project.

Choosing the right consultant is a difficult proposition for many systems companies. Of course, everybody has marketing materials that trumpet their FPGA design experience and expertise, but how do you tell who can actually deliver the goods? For FPGA companies, this is an important problem as well, for two reasons: First, many design teams engage a consultant before they’ve chosen an FPGA. It makes sense. If you are bringing in an FPGA expert, you might as well use their expertise to help choose the right device in the first place. FPGA vendors, therefore, want you to choose a consultant that will recommend their FPGAs. Duh! Second, FPGA companies want you to succeed with your design. If you choose the wrong consultant and your project flounders, the FPGA company won’t make any money. Furthermore, you may tend to blame them for the difficulties in the FPGA portion of your project. 

The bottom line is, FPGA companies want you to choose a consultant that will direct you to their FPGAs and then proceed to drive your design to success with a minimal amount of hassle.

Since about a year ago, Xilinx has maintained a list of consulting firms called the “Alliance Program” that could help you with your design, but recently they have upped the ante with a complete re-vamp of their consulting and IP partner program. The main feature of the revamp is the creation of a new top tier of service providers called “Premier Design Services.” According to Xilinx, firms that have achieved the “Premier” designation have “demonstrated that they can execute to the industry’s most rigorous criteria of technical, business, and quality standards.” Translation: These consulting firms not only have a track record doing Xilinx designs, but they have also successfully jumped through some pretty big qualification hoops.

The Xilinx Alliance program now has three levels: Member, Certified, and Premier. The “member” level is the easiest to achieve, of course – you have to fill out an application and stand up to a certain minimal level of scrutiny on your ability to complete a Xilinx FPGA design. (We suspect you also have to pinky-swear to never, ever recommend switching to an Altera FPGA, but we could find absolutely no reference to this requirement in any Xilinx documentation.) The Certified level is where they turn up the heat. In order to become “Certified,” consulting firms have to complete a corporate self-audit and commit to regular certification training and testing. Both the firm and the individual consultants have to qualify. The training requirement is substantial, so firms achieving “Certified” status have consulting engineers who have spent a good deal of time in Xilinx training classes, and they have done so recently. Today, there are somewhere around thirty firms worldwide that have achieved the Certified designation. 

To reach the new Premier status, the requirements are daunting. You have to name your first three children Kintex, Artix, and Virtex. OK, that’s not true. You do, however, have to start with a track record of successful Xilinx design engagements. You have to have a broad range of technical capabilities that span the gamut of Xilinx offerings. You have to be scalable, with the ability to tackle and boost large projects with adequate staff and engineering bandwidth. Then, you have to submit to a Xilinx on-site audit, including a review of your quality processes. You also have to go through basically the same level of recurrent training as a Xilinx FAE, and you then have access to the same priority hotline and webcase assets as Xilinx FAEs. Finally, you are awarded a spandex jersey with a big “X” logo on the chest. (You guessed it. That last one is a lie as well. The rest are all true, though.)

So far, three firms have achieved the Premier Design Services designation – selected from the thirty or so that are Certified: Tokyo Electron Device – serving Japan and Asia-Pacific, CoreEL Technologies – serving the whole world, and Fidus Systems Inc – serving North America. These firms are expected to be joined by additional consulting houses as more qualification processes are completed.

This kind of care and feeding of the FPGA ecosystem is a relatively new development in the FPGA landscape. While FPGA companies have long given lip service to third-party ecosystem support, the reality was often much less impressive. FPGA companies have a history steeped in NIH (Not Invented Here) syndrome, where it seemed for years like they felt they could do everything themselves, from silicon to software to support and IP. Being a third-party supplier of FPGA-related technology or services was a slippery slope, as you never knew if the FPGA companies were your partners or your competitors. As the market has expanded and the demands have grown, it has been difficult for FPGA companies to change these ingrained behaviors. The Alliance program is a good example of success in this area. Let’s hope many others follow.

3 thoughts on “Help! Help!”

  1. Does anybody know how to join Xilinx Alliance program as a Member?
    Where can I find more detailed information about requirements and obligations?

Leave a Reply

featured blogs
Jul 20, 2024
If you are looking for great technology-related reads, here are some offerings that I cannot recommend highly enough....

featured video

How NV5, NVIDIA, and Cadence Collaboration Optimizes Data Center Efficiency, Performance, and Reliability

Sponsored by Cadence Design Systems

Deploying data centers with AI high-density workloads and ensuring they are capable for anticipated power trends requires insight. Creating a digital twin using the Cadence Reality Digital Twin Platform helped plan the deployment of current workloads and future-proof the investment. Learn about the collaboration between NV5, NVIDIA, and Cadence to optimize data center efficiency, performance, and reliability. 

Click here for more information about Cadence Data Center Solutions

featured chalk talk

Versatile S32G3 Processors for Automotive and Beyond
In this episode of Chalk Talk, Amelia Dalton and Brian Carlson from NXP investigate NXP’s S32G3 vehicle network processors that combine ASIL D safety, hardware security, high-performance real-time and application processing and network acceleration. They explore how these processors support many vehicle needs simultaneously, the specific benefits they bring to autonomous drive and ADAS applications, and how you can get started developing with these processors today.
Jul 24, 2024
972 views