Elastix Clocks Reduce Margins. Landing in Barcelona.

Warning: Undefined array key "upload_with_nude_flag" in /www/www_com/htdocs/get_weekly_feature_ad.inc.php on line 69

Introduction

Elastix is an early-stage EDA company developing innovative technologies to enable variability-aware designs and optimize power-performance trade-offs for 65nm and beyond technologies. Their technology of elastic clocks enables designers to reduce performance and power margins. The technology comes out of the Barcelona area which has an innovative incubator program. I had an opportunity to talk about the technology, the company and the Landing program with CEO Vigyan Singhai and Engineing VP Emre Tuncer.

Would you provide us with a brief biography?
My name is Vigyan Singhai. I am CEO of Elastix. Along with Emre (Emre Tuncer VP of Engineering) and Jordi (Jordi Cortadella, Chief Scientist) I founded the company. I got my Ph.D. degree in logic synthesis and verification at UC Berkley under the guidance of Professor After that I joined Cadence for a few years. After Cadence I founded Jasper Design Automation where I was CEO for the first three years until the day of the second round financing. I stayed at Jasper for a few more years. I left to start a design verification and IC company called Oski Technology. In 2006 I was introduced to the idea of elastic clocks by Jordi Cortadella who is our CTO and Luciano Lavagno. These two guys are the inventors of the technology. The idea was very intriguing to me. It promised a lot of performance and power gains. I did a lot of due diligence, tried to understand the potential of the technology as well as the market opportunity by talking to a few potential customers. Once I was convinced I left my business to jump into this full time at the end of 2006.

A brief biography from you also, Emre.
I started my professional career in EDA in southern California. After finishing my Ph.D. at the University of Texas at Austin I moved to Camarillo and joined Quad Design. At the time it was a subsidiary of View Logic. I was there for two years, mostly on signal integrity. We developed a prototype of our own chip signal integrity tools and for the customers too. About two years later I moved to the Bay Area to Monterey Design Systems. At the end of 2002 I left Monterey and went to Magma. My first responsibility at Magma was for engineering. In the subsequent years I held VP of Engineering and Chief Technologist positions. About a year ago I joined Elastix. I have been working on the product ever since.

Emre: What caused you to leave Magma which has been doing very well to join a new startup?
I got the startup fever when I joined Monterey. I was one of the initial people there. Then I joined Magma. I saw Magma growing up from when I joined it was around 100 to 200 people to over 800 people when I left. What I wanted to do was take a new idea, a new technology, from scratch and bring it to the user domain and make it a product. I met with Vigyan. After numerous discussions I realized that this was the right technology. It is the right technology in the sense that it is not disruptive but the benefits it provides are much higher than the things you need to adopt the technology.

How did you (Vigyan) learn about the technology?
I was introduced to this by Luciano Lavagno who I have worked with for many, many years. He got his Ph.D. at UC Berkley, was at Cadence and is also Professor at Polytechnico de Torino… right now. He was one of the inventors of this technology. He introduced this idea to me a couple of years ago. Since then I have been working on this. These guys have been working on this technology for about 10 years now.

Where were they working?
At universities. Luciano Professor at Politecnico de Torino in Italy and Jordi Professor at Barcelona. They also had a couple of other researchers. One guy was from Cadence and worked at Philips.

Did you bankroll the company or did you get outside funding?
Initially we funded the company on sweat labor. Then we got a big investment from an angle investor last year. Today we are raising additional funds.

If I understand correctly Elastix has one foot in Silicon Valley and one foot in Barcelona.
The city of Barcelona had adopted a lot of initiatives to bring in new companies. There is a section inside Barcelona, called the 22nd district, which is a vision of creating a compact city where one can live, work and play. It is kind of like a sub-city inside Barcelona. They have the Landing Program which provides an infrastructure to get a firm up to speed in all the issues such as legal, accounting, employment, and getting office space. We had no experience before in starting and building a business in Barcelona. The partnership with that program has been extremely kind and fruitful for us because without their help for us to go there and start this business would have been very difficult. It is an ideal environment to have from the perspective of a foreign company which is landing in Barcelona. The district itself envisions itself becoming a business landing zone for international companies. It pretty much represents what Elastix is. We are in Barcelona because Jordi has been there for many years. He has a lot of contacts, academic contacts; his ex-students, resource into a pool of people we can tap into. But we did not have the business contacts. For that the Landing Program has been extremely helpful.

Are there any other EDA or semiconductor companies in Barcelona?
That is the interesting thing about Barcelona. The economy is strong there in a lot of industries like telecommunications, oil and gas, finance, electronics and high tech. There has not been a concentration of large companies but new companies are coming up. There are semiconductor companies like DS2 which is a well funded company. Barcelona Design Systems was named because the founder was from the city of Barcelona but it had no operations there. We can take advantage of the talent right in the city itself.
Editor: Dr. Mar Hershenson was a founder of Barcelona Design. More recently she was CEO of Sabio Labs which was recently acquired by Magma.

What was the problem that the technology was trying to solve?
The way I look at it, we had a vision of building a large complex chip without sacrificing significant performance and power margins. What has been happening for many, many years is that we were throwing away large margins because the only way we knew to do design was to do worse case design, sacrificing significant variations in margins from the typical design points. Today in the face of increasing low power requirements we are forced to deal with variability. With our technology we embrace variability and let the design run at their natural power and performance which is significantly lower than how chips are designed and manufactured today. That’s the opportunity. People are doing all sorts of things to manage power, control power, minimize power requirements both in the design of blocks as well as in the design of interconnect. There is a rise in adoption of aggressive voltage scaling techniques. That is prompting people to think about variation, off chip variation, timing analysis and physical design. It was the right time for the technology to come about. It was a natural and easy answer to a lot of those issues.

What is the source of this variability?
Variability in design is there. Some people say it is increasing, others say it is the same. The sources are different. There are the obvious manufacturing sources. Different devices on the same chip may have variations for manufacturing reasons. On the same wafer they may have different performance. Of course different wafers form the same fab or from different fabs, even with exactly the same design and process and so on. Similarly there are environmental variations because of changes in supply voltage, temperature. In addition to all that there are also data variations. Under different logical input conditions the same design or the same block may run in different amounts of time. We can take advantage of all those variations. You do not have to worse-case all the variations all of the time and sacrifice or throw away significant timing or power.

What is the approach that Elastix is using?
Our technology is based upon the concept of elastic clocks. We implement the global clocking network in the design so that the clocks become elastic and flexible in that it tracks the performance of the design. If the data path happens to be fast because the temperature is low, if it happens to be slow because it is on the slow part of the chip or because the voltage is low for some period, then the clock automatically tracks the data path. The clock is slow or fast depending upon the data path. By virtue of the clock tracking the data path, I do not have to make the clock slower than the worst the data can ever be.

So it is dynamically doing this?
It is happening live, dynamically. The advantage of all of this is that you can take any average block in the design, change the voltage of that block or any portion of that block. The design is still guaranteed to work but the performance is going to resonate with how you play with the voltage for example. All this is happening dynamically.

When you are verifying the chip, how do you know that this is in fact what is going to happen?
Good question. There are various verification aspects. One is verifying the design itself, that the design works and is correct. The nice thing about our approach is that it fits into existing design flows. We use the same standard RTL to GDSII flows, the same place and route, the same timing and functional verification. It is exactly the same RTL verification. That has not been touched. There is a problem of functional verification. We have altered the design. We have introduced some new circuit elements into the design. You have to verify that. The original circuit is a thing of the past. You use functional verification and certain tools from companies like Cadence and Synopsys. The results from timing verification is to verify that indeed at the corners the design still works and provides the performance you want. That is verified through sign-off, multicorner sign-off using standard timing verification tools, signoffs tools from Synopsys for example. The signoff is done at all the corners of the process, voltage and temperature.

Does the design get to some point of verification and then does your technology enter the picture? Is it a sequential process? The designer does some design or some design block, verifies that design, arrives at some comfort level and then brings in your technology (elastic clocks) and re-verifies.
Yeah, kind of. You have some design in your mind, some conceptual level, some system level. At some point you write the RTL for that design. You go through standard verification using whatever simulator you have. We do not come in until later. Then you synthesize the design using Cadence, Synopsys or Magma tools. Once the design is synthesized and is at the gate level, you are inserting the clock tree. That is where we come in. You apply the Elastix transformation. After applying that you get back into the standard design flow, physical synthesis and routing. You can do the verification immediately after applying the Elastic transformation as well as at all the subsequent stages.

We start from a pre-verified design and then we make any changes through our tools. We re-verify the impact of those changes though timing and functional verification.

Do you have any customers or beta sites for this technology?
Yes there are customers. We are working with some strategic customers who are checking out their designs. I am not in a position to disclose their names but we are working with several customers.

Have you set a potential price point for the product?
No. Actually we have not released the details of our product except to some customers. This will come in a few months.

Is there a particular set of end user applications for which this technology would have maximum benefit?
First it works for all designs but the sweet spot for this technology are designs which are system-on-chip where potentially you will be putting together blocks from different sources, from inside the company or from third parties. A lot of the time the devices are working for multiple applications. The chip that is in your cell phone is sometimes used for videos, sometimes for email and some times for simply talking on the phone. So you have different workload requirements. There are different voltage requirements under different workloads. Different blocks scale and perform differently. You need to be able to support different power and voltage requirements when blocks behave differently. So that is the perfect sweet spot for elastic clocks.

Another sweet spot is multicore applications where people are building multiple ports putting them together on the same chip whether it is a graphics chip, a simple process or router chip. They are putting many different ports onto the same chip and connecting them together. Any applications where you are putting multiple blocks together designed under different conditions and requirements and having to talk with each other and be interconnected and you have some flexibility in the connect in terms of the work requirements and variation. So low power applications.

How large a company is Elastix today?
We are about 10 people.

Do you envision selling directly or through distributors?
It depends upon the market. Initially we will be selling directly but as we go to Asia we will work with distributors.

Do you have a time frame for releasing the product?
In a few months.

Possibly at DAC?
No. We will not release before DAC. We are working with customers who have our product and are using it on their designs. We do not have an availability date at this time.

To your knowledge is there any other firm out there doing something similar?
There has been a lot of research in voltage scaling as well as making clocks flexible or less rigid. In terms of voltage scaling techniques there are a lot of in-house solutions in different places. They have to do with adaptive voltage scaling as well as dynamic voltage or frequency scaling where they are changing the voltage as the frequency of the block changes. But all of them are in the context of rigid clocks. Then in terms of changing clocks, there have been a lot of work in different segments. In the context of synchronicity and rigid clocks there have been efforts in implementing useful skew. There have been efforts in taking into account of on chip variation also called OCV. Then what we are doing is making the clock even more flexible than all of that. There are other companies that have taken radically different approaches like Handshake Solutions. Our approach is significantly different than that.

Why is your approach more effective than the more popular or traditional voltage scaling approach?
It is not that we are competing with any of the voltage scaling approaches. The issue is that we are able to exploit the voltage scaling. We can make voltage scaling as flexible as you want. You can make it at as fine a granularity as you want to make it. With voltage scaling today you are limited to scaling the voltage and scaling the frequency and doing some tight synchronization between the two steps but in doing that you are losing a lot of margin and you have to go back to voltage specific areas. In our technology you can scale voltage independently on any portion of the design or any sub-block you choose to. It is the most flexible way of voltage scaling.

Is it conceivable that a design could be so poor that you could not make it work, close timing even with elastic clocks?
I can not think of something like that. Are you saying that the design can not be made to work even in a synchronized rigid clock world?

Let me rephrase the question. Is it a prerequisite to using your technology that a design works first in a rigid clock world?
Yeah. It depends on what you mean. You can have designs with rigid clocks that do not meet your timing or power requirements and by using our technology you can get the design to meet those requirements. So a design that was not working in some sense is now working. If you had a design that was working, it will always work because with elastic clocks you can get additional benefits. You have to use the tool and get the results from the tool and the clocks become elastic. But if the design was working, it will always work at least at the same speed and power as before but typically much faster and at lower power than before.

Anything else to add?
Why are we doing this? Power has become the number one design issue today. People are doing designs with more modes of operations, voltage scaling and so on. We have an opportunity to make these designs as flexible and as of as low power as possible. That’s the opportunity we have. We are excited about ding this.

The most significant thing is to make all of this work in the different existing design flows, with the existing sign-off tools, timing analysis and functional verification. That is something different in what we are doing as opposed to previous solutions in the same space. We are excited about working with our strategic customers and making this happen.

By the virtue of clocks being elastic, the designs run faster, and for power-sensitive applications, the same performance can be achieved by running the design at lower voltages, thereby saving both dynamic and static power. A qualitative illustration is shown below.

Editor: Barcelona’s Landing is a high value strategic program that grants knowledge-intensive and technological business initiatives the access to a network of platforms in different Innovation Systems making their internationalization easier and quicker.

Landing offers a privileged platform: an international incubator at Barcelona’s Innovation pole, one of the biggest European metropolises which is also the economic, cultural and administrative capital of Catalonia, the region located at the northeast of Spain, in the Mediterranean sea.

22@Barcelona is a project that is transforming two hundred hectares of industrial land of Poblenou into an innovative district offering modern spaces for the strategic concentration of intensive knowledge-based activities. The project is being designed to become a ‘compact city’. Having the Triple Helix as a model, it creates an environment and culture propitious to development and innovation relating The Government, University and Industry, where, each of the actors will both bring value to the Program and at the same time be the direct, priority recipient of the value proposition offered by the Landing Program. Finally, 22@Barcelona is fostering the development of technology clusters within the district, specifically ICT, MEDIA, MEDTECH and ENERGY. The infrastructure is being deployed, organizations and agents with research capabilities are settling in the district, and if territorial innovative capacity is displayed, as a virtuous circle, entrepreneurship and talent will be called and attracted.