Monthly Archives: September 2014

ip cycle

Defending the IP of the US Semiconductor Industry

This is a guest post by Apek Mulay that was originally publish on EBN.  


Dramatic political and economic shifts are common in the Middle East. In addition to profound impact on those regions, these changes also inform the path forward for the US semiconductor industry. It is very important to understand macrohistory to help analyze the impacts of geo-political events on the operation of businesses.

 

Today, the US semiconductor industry has a huge vested interest in the Middle East. According to the Census of foreign trade, United States firms exported nearly $24.81 billion worth of goods to the United Arab Emirates (UAE).

 

Further, leading semiconductor companies are developing deep ties with the region. For example, well-known full-service semiconductor foundry GlobalFoundries is owned by Advanced Technology Investment Company (ATIC). ATIC, meanwhile, is owned by Mubadala Development Company, a wholly owned investment vehicle of the Government of Abu Dhabi, in the UAE. The ATIC is also planning to set-up a 300 mm fab in Abu Dhabi in near future.

 

As ties between the UAE and the US semiconductor industry develop, we must remain aware of the potential implications of these shifts.

 

Applying a framework 
An idea from writings on macrohistory by an Indian philosopher and spiritual leader provides a helpful framework for this discussion. The Law of Social Cycle has its source in the concepts of Macrohistory presented in Shrii Prabhat Ranjan Sarkar‘s philosophical treatise, titled Ananda Sutram, along with original concepts of metaphysics, epistemology, and ethics.

 

 

Figure 1: Law of Social Cycle explained in pictorial form. As shown in figure above, control of society keeps moving from intellectuals to acquisitors to laborers to warriors in a clockwise or anti-clockwise direction depending on the domination of intellectuals, acquisitors, warriors, or laborers.
Figure 1: Law of Social Cycle explained in pictorial form. As shown in figure above, control of society keeps moving from intellectuals to acquisitors to laborers to warriors in a clockwise or anti-clockwise direction depending on the domination of intellectuals, acquisitors, warriors, or laborers.

 

 

According to Sarkar’s writings on macrohistory, in this movement of the social cycle, one class is always dominant. The movement of the social cycle in a clockwise direction in Fig 1 (shown by blue arrows) constitutes an “evolution” if it occurs after a sufficiently long duration. If this clockwise movement occurs within a short duration, this is called “revolution.” The movement of the social cycle counter-clockwise in Fig 1 (shown by orange arrows) constitutes a “counter evolution” if it occurs after a significantly long duration. This counter-evolution is extremely short-lived. However, if this anti-clockwise movement occurs within a short duration, it is called “counter revolution.” Counter-revolution is even more short-lived than counter-evolution.

 

Professor Ravi Batra from SMU, Dallas, analyzes the successful operation of the Law of Social Cycle in his 1978 book The Downfall of Capitalism and Communism: A New Study of History. Based on this analysis, Professor Batra correctly predicted the collapse of Soviet Communism 15 years before it occurred.

 

The above concept of Macrohistory could be used to analyze the recent social and political events like the Arab Spring in the Middle East. Based on this law, as shown in Figure 2 below, the rise of fundamentalism would have grave consequences on the US semiconductor industry because the industry relies on its investments coming from the Middle East.

 

 

Figure 2: Recent Social events in Egypt complying with Shrii P.R. Sarkar's Law of Social Cycle showing different stages of Evolutions and Counter Revolution.
Figure 2: Recent Social events in Egypt complying with Shrii P.R. Sarkar’s Law of Social Cycle showing different stages of Evolutions and Counter Revolution.

 

 

We have discussed in the past how the US may be accused of transferring economic dominance to China through its trade and monetary policies. At the same time, it is clear that the UAE is also growing its economic ties with China.

 

The inability of the US congress to pass a balanced budget has forced UAE to grow its economic and political links with China and China has also signed a currency swap deal with the UAE. The growing ties between China and Abu Dhabi will influence Abu Dhabi’s decisions in the best interest of China, which may not be in the best interest of the US.

 

Abu Dhabi’s ownership of GlobalFoundries in New York will act as leverage in the transfer of advanced semiconductor manufacturing technology from its fab in New York to its upcoming fab in Abu Dhabi. Growing ties between the UAE and China could also transfer this technology to China in the future as the UAE is no way financially dependent on US.

 

Taking into consideration the threat of transfer of technology to China, the US needs to undertake major reforms to have a balanced economy. Additionally, while the US should support growth of education and infrastructure in UAE, it must also strive to ensure that the intellectual property (IP) of its semiconductor industry is protected.

 

Hence, a top-notch fab like GlobalFoundries should become financially independent of any foreign investments coming from Middle East. As a first step, wafer fabs should develop symbiotic partnerships with the US Government to ensure the sustainability of its capital investments. (Read more here.)

 

Let us know what you think about the smartest path forward for US semiconductor manufacturers in the comments section below.

methodology

The Right Process For Verification IP Development

This is a guest post by Arrow Devices, that provides high-quality Design & Verification products and services for ASIC/SOC. 


Verification plays a major role in any chip design project. Within that, functional verification takes lion’s share of any design cycle, 70% by some estimates. Functional verification completeness is extremely critical. Following the wrong process for developing verification IP can cause you to miss verifying some features. This can cause chip failure and cost the company millions of dollars. It has been noted that close to 70% of chip re-spins are typically due to functional bugs although that percentage has come down to ~50% in recent times. Thus following a good process with lots of planning is essential for development of world class verification IP.

 

Don’t “Methodologies” already exist?

Standard verification methodologies such as VMM, OVM and UVM mostly focus on the test bench building methodology rather than the process of building a complete Verification IP. Although there have been attempts to standardize the process of the building Verification IPs but they have been in there early stages. Hence there is need for a more complete methodology approach towards building Verification IPs.

 

So what’s the right process for developing a Verification IP? Is there any?

Yes there is! And we call it CVM – Comprehensive Verification Methodology. So here we are sharing with you what we think is the right process one should follow to come up with best-in-class Verification IPs

 

Continue reading the article on Arrow Devices blog

find asic vendors

Can you reduce cost and development cycle of your next ASIC?

Have you ever wondered about the possibility of minimizing risk, design time and production cost simply by working with different IC design or verification companies? Or whether you could make your current ASIC cheaper by changing your FAT (foundry, assembly, test) providers?

 

Actually it’s something you can do right now.

 

We have released today a new and improved site focused on helping ASIC designs and operation managers find semiconductor service providers with the best fit for their next design. The new site provides new capabilities, usability enhancements and significant speed improvements.

 

Our new site helps visitors search quickly for providers using categories like IC design companies, verification, packaging, qualification, testing and failure analysis.  Selecting the right supplier with the best match for your project will essentially reduce your time to market, minimize risk and ultimately reduce development costs.

 

We have improved our search so that it’s accessible from every page and offers criteria selection to help you narrow down the results. If you need additional help, we now offer a new free-of-charge service called Concierge. Via the Concierge page you can fill out a quick information form, which our team analyzes and then pushes the requests to the right service providers. These providers can then contact you back directly.

 

Every vendor page consists of a “contact” button that lets you send direct email to the most relevant people at the service provider.  This provides you with immediate access to the correct people.

 

What else? Our semiconductor blog is now open to external contributors – both individuals and companies. So if you’re interested in sharing your knowledge and provide practical hands-on tips on topics such as IC , design, layout, wafer, test, packaging, please contact us using the  “Submit an article” link and drop us a line with information on the article.

 

Last but not least, our blog does not require any registration and is open to all visitors. Visitors are welcome to use our new comments platform and join the discussion.

 

Love and lots of passion,

The AnySilicon Team.

donts

Don’t Get Critical IP Cores from a ‘Supermarket’

This is a guest post by PLDA which designs and sells intellectual property (IP) cores and prototyping tools for ASIC and FPGA


You are on a tight schedule for your next chip. Not wanting to reinvent the wheel, you plan to go to an outside vendor for some of your silicon IP core. So now you face the decision do I get my IP core from a large supplier who offers many cores covering a broad range of functionality (the supermarket) or do I go to a focused vendor who specializes in the critical piece of IP core I need (the specialty shop)? Well, it depends on the IP core.

 

When you need a particular piece of third-party IP core, the factors that go into the where to purchase decision include the hard attributes of IP functionality, availability for a particular process node, performance, complexity, ease of implementation, testability, and security. Along with these attributes are others that apply specifically to vendor selection, such as track record as an IP supplier and level of service. The latter is particularly important if problems arise during IP integration. An IP supplier who has been in the business for several years and is profitable demonstrates that they are in for the long haul and will still be around down the road if problems develop or to help with tricky IP integration technical problems on future chips.

 

A small IP vendor may be your best choice if you need a specialized IP core. Since a small vendor has fewer products than a large vendor, they often are more willing to work with you to make sure their product works in your particular chip. In other words, small vendors generally provide better support and extra service to their customers. Treat your IP vendor as a member of your design team to help solve any and all problems relating to the IP from that supplier and go with one willing to accept this role.

 

A real test of an IP vendors value is what happens when you implement IP and it doesnt work correctly. A vendors knowledge needs to extend beyond the actual IP that they sell, including knowledge of the various types of systems in which the IP may be embedded. Look for a vendor who has experience with several types of designs using a particular piece of IP.

 

An important factor beyond the particular core you want is how involved the vendor is with similar IP. For example, if you need PCI Express (PCIe) IP does the vendor also have PCI and PCI-X experience and knowledge? Experience with similar functionality in other cores indicates that the IP vendor most likely has extensive familiarity with the core functionality you need.

 

Parameterizable IP is another plus, allowing you to customize one core for a basic design and variations of the same IP for derivative products. As is the case with any semiconductor product, find out what else the vendor can supply beyond the basic IP. If the vendor has versions of a particular core for different silicon platforms ASIC, FPGA and maybe even Structured ASIC this gives you flexibility during product development (maybe on an FPGA) and then for transitioning to an ASIC for cost advantages. If your critical need is for IP to implement a communication standard, such as PCIe, then your vendor should offer IP verification with common verification IP (VIP) that you might employ, such as that offered by nSys, along with verification reports. Depending on the IP, the vendor should also have the appropriate software youll need to implement the IP in a system.

 

Reference designs from the IP vendor, particularly when you are targeting an FPGA platform for design validation or prototyping, will save you a lot of design time. A small, experienced IP vendor will generally be a good choice because they have the design experience to develop such boards with high performance characteristics and ease of use.

 

The bottom line is to look beyond the IP itself and see what else the vendor has to make system implementation and verification as painless as possible. IP integration is not turnkey for high-performance, critical IP and it never will be since every chip in which the IP will reside represents a different system around it. A small, focused IP vendor is your best bet to get the core, associated hardware and software, and service you need to maximize successful IP integration.

 

See more from PLDA click here.