First of all, we'd like to welcome all of you in the room and also welcome those of you on the webcast, and thank you for taking the time out of your day to join us. I suppose I should do this. Okay. Before we begin, the safe harbor statement. We will be making forward-looking statements today, and we encourage you to review our annual and quarterly SEC filings for a list of risks and uncertainties that could cause actual results to differ materially. Now to the good stuff. Today, we have three executives with us. We're going to start out with Chairman and Co-CEO, Aart de Geus. Aart will set the stage by discussing our growth strategy in the context of the many evolving technical challenges facing our customers today and tomorrow.
Also discuss the Synopsys strategy for attacking those technical challenges on behalf of our customers. We will then have Dave DeMaria, who's Vice President of Corporate Marketing, provide some highlights from some of our highest impact, highest profile products. Lastly, you'll hear from Andreas Kuehlmann, who is Senior Vice President and GM of our Software Integrity Group. About a year and a half ago, we acquired a company called Coverity and entered a brand-new higher growth TAM. Andreas was head of R&D for Coverity, and we're very happy to have him as part of our executive team. Since then, we've expanded our presence and our opportunity space more in the security arena. Andreas will go into some depth today describing the challenges driving the emerging market space of software quality and security.
He'll also do a bit of a deep dive in the security space, and he'll update you on the Synopsys strategy and also operations. After that, we'll bring all three up here, have a Q&A session, and we will conclude by 1:00 P.M. With that, I will hand it over to Aart de Geus.
Well, thank you.
To Aart de Geus.
Well, thank you, and good morning. Thank you for attending this session. The objective of the session is to give you a sense and feel for the company in its entirety and direction, especially in light of unbelievably rapidly changing markets, technology. Almost a phase shift of what has happened in the silicon world. In order to do that, I'd like to give you a little bit of a sense about some of the dynamics that drive us and how we have structured the strategy and the execution of the company against that. For starters, obviously, our pedigree is deep in silicon. Now we're almost 30 years in that. You may know that this year was 50 years of Moore's Law. Moore's Law was this governing driver towards smaller transistors, faster transistors, lower power transistors. By the way, that drive is continuing.
The part of Moore's Law that's a little bit in question is, will they also continually get cheaper? So far, people are moving very rapidly in that direction. What is interesting to me, and really already brings you immediately to the heart of what I think our strategy and our communication to you is about, is that with this continued drive towards still more capabilities in silicon at lower power, lower footprint for footprint, we're entering a phase where the software world will see yet another massive expansion of opportunities. In many ways, the center of gravity of this picture is where we are gradually moving the company.
We're doing that in a fashion that hopefully is coherent, where you can see that they're both business opportunities that line up, but also the technical acumen of the company coming up from the silicon applies completely to the software world. By the way, we ourselves are one of the largest software companies in the world and certainly one of the most sophisticated ones, with about 400 million lines of code ourselves. This is not your grandmother's code. This is super sophisticated stuff. There's plenty of opportunities to have issues ourselves, and therefore, we're very much attuned to how do you do that well. If you look at Synopsys, and this is not meant to be a financial presentation, but it's still useful to look at it in context. Earlier, I had a question of, what happened in 2009?
Are there possibilities that would happen again? Well, you can actually see on the picture 2009, and you can see, yes, it was a little flatter for us, but we actually navigated through this extremely well. The reason for that is that fundamentally, the technology needs that we serve continue with or without downturns. That drive is there. Secondly, our business model of a reasonable multi-year engagement and commitment from customers has served us extremely well in these challenging times. Third, when there are consolidations, more often than not, customers tend to look at the leader, be it in technology or in market share, as the most reliable, trustworthy partner. We've benefited from that. To give a little sense of scale, today we're roughly at a $2.2 billion size of the company. In our industry, we are certainly the leader in the EDA side.
We are number two in the IP space, we are now, as you know, have entered the software space. The business model has been very solid and coherent for a long time, with good cash generation, fairly high degree of predictability. Our objective, of course, is to manage all of these technology transitions, market transitions, while over the long term, consistently deliver shareholder value, and we anchor that mostly in the drive towards EPS growth. In order to do that, we have seen the opportunity to diversify from where we were, because once you become a leader in a segment, it's time to look at sideways segments. Of course, we did that already for a couple of decades.
If you look at the EDA segment, we are the market leader there, most importantly, our pedigree, and it remains absolutely that, is lead and drive the technology with our key partners, mostly in the foundry space and the most advanced design companies. Those are high-demand partners, and that's a good thing because it drives us forward. We have a number of new products, and Dave will be talking a bit about what we're doing there. Secondly, as you know, since the early 2000s, we have really built a substantial position in IP where we're number two behind Arm, which is a great company providing the apps processors. We provide essentially everything else. We have far and away the broadest portfolio, and we have a lot of experience in working with many customers in providing IP for them.
Most recently, meaning the last year and a half, you heard Lisa say that we entered the software world, which is a little bit of an open-ended statement. We like to call it Software Integrity because it captures everything from quality to testing to security and now increasingly also safety and other questions. Here we see a growing TAM, a TAM that's getting a lot of attention. It's quite fractured, but there's an opportunity for us to bring coherence, build a platform, and align technologies that are very fundamental and will be needed by everybody over time. If we look at this picture, it sits very much in the middle of these massive trends.
Of course, if you look at the silicon industry, its historical phases of the PC that became servers, became compute centers, then the phone that became mobility, that became smartphone and pads and so on. The question is, so what's the next big thing? Of course, everybody says the next big thing is sitting in the middle. It's IoT, Internet of Things, then they all think something different. Nothing wrong with that. That is the characteristic of an early-phase market, which is a lot of creativity, let's see where the big money is. Whenever one cannot predict where the big winners are going to be, we always think, what are the fundamentals that govern this? If there's any fundamental that I'd like to highlight is that I think this is as much as anything the new intersection between hardware and software.
Of course, we come initially from the hardware side, but we know software very well, many of our initial customers, the semiconductor guys, actually today have more software engineers than hardware engineers. They already provide this. Of course, the software world moves all the way to high-end apps, be they in the financial domain or the health domain or cars, you name it. If we look at this IoT world what one will see is that very soon we will see little IoT things appearing everywhere. Of course, this can be very simple things such as smart thermometers that you have in the wall and so on. It can be very sophisticated things, facial recognition, feeling recognition, things that suddenly gives a very new insight. You can highlight automotive as a domain that's bringing smarts everywhere.
The other characteristic, of course, is that all of this is connected in many ways, I sort of like this picture because it's a little bit a take off of the human brain with a lot of little neurons talking well to each other. This is the essence of what people refer to as big data, which is you get all this data. Now do something interesting with it. Now, of course, you have the application domain where the smarts of the algorithms, the business models around that determine the value. Be that as it may, this is often referred to as the Internet of Things, IoT for short, which at times I like to call incredibly optimistic thinking. On one hand, that's paraphrasing a certain degree of skepticism about the opportunity space. On the other hand, I do believe in the optimism here.
This is very fundamentally changing the world, we hear mostly about the stories that start sort of on the edge of things, the devices that pick up through sensors, some physical data in some form or another. Of course, those devices immediately then have to connect to the cloud, no matter what, therefore the cloud will continue to grow, be it in computational capabilities, in the transport, in the bandwidth available, in the memory. These things will be insatiable as suddenly millions of devices start to create little data. The reality is the edge rarely connects directly to the cloud. There's often what we affectionately call the fog, the in-between, which is, you have a piece of fog in your pocket right now. It's your phone.
It's a local aggregator that can maybe talk to your Fitbit, that will talk over time to some health devices on your body and so on. The same is true in the house, it's true in companies. We'll see essentially networks of networks, some being internet, some being localized. The picture definitely will develop in a very hierarchical fashion. On top of that, we should not forget the apps, because that is, in many cases, where the money enters the system. This is the fundamental stack that we're talking about, and it's very hardware-oriented towards the bottom. It's completely software at the top. Even at the bottom, we will see a rapid increase of the amount of software. Another perspective on exactly the same story is to say, well, look at it from the end market.
Of course, we're all familiar with many of those. The wearables is the most visible one because it's sort of consumer stuff, it's the fun stuff and so on. It is very clear that these other domains have tremendous increases in sophistication and capabilities. One of the ones that has now become super visible and now suddenly also super competitive, take automotive. The amount of sensors and the amount of software that is in an advanced car today is quite amazing. Someone was saying that the amount of silicon is the equivalent of a complete 300 mm wafer of silicon in chips, which is a lot of chips. By the way, 100 million lines of code.
Now we're entering the assisted drive, automatic drive age, and you can just see the number of sensors, the number of ways of looking what's happening around, but most extraordinarily, the demand on computation inside of the car to essentially try to mimic what a human is doing is unbelievable. With that, of course, come all kinds of challenges and opportunities, and my favorite story is Tesla introducing or downloading via software the capabilities to automatically park your car in a parking garage. I think they're doing that this summer or so. Question I have for you, which is, what happens when two Teslas simultaneously see the same open parking spot? Software, yeah. Conflict mitigation software in the car.
There's a zillion of these fun stories, but fundamentally, this landscape is changing everywhere, and each one of those industries will find its spot where the impact has the highest return on the economics. It is very difficult to predict which ones will be fast, the winners, and so on, but it is absolutely clear they're all facing the same challenges, this interaction between hardware and software and the applications that finally run on this. With that comes, of course, also some other challenges, which is that if you make everything smart and everything's connected, there's some other guys that figured this one out, too. You're very well aware that literally every week you see another breach in some form or another situation being hacked.
The very fact that just in the last couple of months, the automotive industry was hammered in unexpected ways here just shows the gravitas of this issue. For a while, I've made the joke that if in your home you use many of these, sooner or later, the police blotter will read, well, the perp came in through the toaster. Not that simplistic a joke as you may think, because Target, when they were hacked, the perp came in through the air conditioning system, which of course is connected to some computer, which is connected to, so it looks like the credit cards. This is the issue we're dealing with, and one of the reasons we're investing in the security angle of software is not because there is a single solution that touches everything. It's very fractured. There are many things.
Again, our objective is to go to the basics of those things that everybody should be doing, and that is our core competence, and I think we're on a very good track there. In summary, if you look at how we look at ourselves is now very much in the center of the silicon to software opportunity space, which is an opportunity space for the next 10- 20 years. In that, we're very cognizant that the drive down the design into silicon is one of our core competencies that we will keep investing in, and that is a big economic engine for us. In parallel to that is, I look at as the drive up, which is verify that what you're building actually works.
That is an area of our space that is actually seeing dramatic increases in problems, especially because the verification is not hardware. It's hardware and software now together. We have made fabulous advances there. Of course, connecting well to the silicon is a big technical challenge. Again, difficult things are good for Synopsys. That's our core competence. Our IP business today provides IP that is connected directly to the most advanced silicon technologies in the world. With other words, that's a risk reducer for the design community that they know with Synopsys they can get to a good silicon. Last but not least, now the Software Integrity business is this new opportunity space that looks not only at connecting well to the silicon, but most importantly, at the quality, the testing, the security, and other aspects of the software.
That's the essence of where we see things going. I prefer to not use Internet of Things. I prefer much more to say smart everything, because by the time you say smart everything, it's very clear that over time, there will be distributed computation needed in every device of any substantial value. That means there will be chips with some software embedded in some form or another. That is the overall market driver that we are aligning behind. With that, let me pass it on to Dave, who will zoom in a little bit more into some of the products.
Thanks, guys. Let me start with talking about IP. IP, clearly one of the big growth drivers is what Aart just described around IoT. I won't call it IoT anymore, I'll call it smart everything. If you take the world of smart everything, there's going to be millions and millions of these devices, and there's going to be hundreds and hundreds of different types of devices. If you net it all out, many of them will share some very common building blocks. Right? Many of them will have sensors that will connect to the real world that we live in. They'll have a processor, and in most cases, they'll have multiple processors. They'll need memory and be driven by software apps. They'll need RF devices for connecting to Wi-Fi and Bluetooth, some kind of power source, and of course, absolutely critical, security will be built in.
In June of this past year, we introduced a new IoT platform built around our IP. Basically, it features the broadest portfolio of IP in the industry and many of the key building blocks that customers need to create these types of devices. Our sensor subsystem, ARC processors, memory libraries, and a whole host of different interface modules that customers will need on their SoCs. Since that announcement in June, we've had several new extensions to the platform. Just last week, we announced some new ARC cores that are targeted at DSP-intensive applications that are very common on IoT, things like sensor fusion, voice recognition, things like video processing, and all of those types of applications. Earlier this year, we announced some key acquisitions. We acquired Bluetooth IP from a company called Silicon Vision, and we acquired Elliptic Technologies, which provide a security IP.
Both of these are very, very critical building blocks for any IoT device out there. In June, we also announced a collaboration with TSMC to pre-validate this IP platform on their 40 nm ultra-low process. They're targeting this for IoT devices. This pre-validation really accelerates our customers' time to market. We're doing the same thing in automotive. We're creating an automotive IP platform that, again, we launched in June. It has many of the same key building blocks I just showed you in IP, but many of the interfaces that are quite different. Ethernet in the car is a big thing, so that's a big component of our automotive IP solution. Probably the biggest difference is in automotive, the big concern is safety. What we've done is we've pre-qualified our IP to a really important standard in automotive called ISO 26262.
We have the safety enhancement package that goes around our IP. What that does is it basically allows our customers to streamline getting their own SoCs qualified for automotive. There's an automotive safety integrity level standard out there called ASIL, and we provide ASIL-level qualification for our automotive platform. This is a really big deal for customers. I think this whole move towards creating these IP platforms will take our IP business to the next level. It'll take our customers' productivity to the next level. Moving to some great progress we've made with IC Compiler II. One of the key drivers is the rapid move to FinFET. Many of you have seen over the years versions of this slide. We're very intimately involved with all of the leading-edge customers, and we track the first 500 designs for each new process node.
What we're seeing is a pretty fast ramp now to FinFET. At last quarter's count, we were at 240, much of it at 16, 14 nm, but you can see 10 nm down the bottom starting to ramp as well. Synopsys is involved as the primary tool in over 95%, 95% of those designs. We're seeing some great momentum. This is one of the key drivers on our growth with IC Compiler II. As a reminder, March 2014, we launched IC Compiler II, it truly is one of the most impactful innovations in our history, and really in the history of EDA, for that matter. It was the result of a five-year R&D effort, which had three main focus areas.
First, we developed a brand-new infrastructure from the ground up, second, we developed some new engines in key areas like floor planning and optimization. Third, we leveraged the best of IC Compiler 1. At the time, and still when we launched this, IC Compiler 1 was the leading place-and-route technology. We started from a position of strength. We had this development process going on in parallel, the goal was bring a 10x improvement to productivity. Not 10%, not 50%, but 10x. We've been seeing some great customer results where they're achieving and in some cases exceeding that. Before I get into those, which I will in a moment, I thought I'd talk a little bit about the competitive landscape, because there's been a lot of noise recently around that.
As I mentioned, before we launched IC Compiler II, IC Compiler 1 was the leading solution in the market by far. We brought out IC Compiler II, which gave a very impressive 10x performance boost over that. About a year later, our competitor launched their new tool, which brought, of course, advantages over their previous system, we still feel very confident we've got a compelling technical advantage, market advantage, and time-to-market advantage. That's really evidenced by the kinds of results that we're seeing in customers. We're seeing a growing number of customers, yet really game-changing results. Some of those shared their experiences at the Design Automation Conference in June. Every year, we host a luncheon where customers share their experiences, I'll share a few of those with you today. We had MediaTek, who were able to meet their very aggressive low-power and schedule goals.
They saw a 5x improvement in speed up. What really that meant for them was they could do their whole design in a day, that was a game-changer for them and allowed them to meet their goals, get their product out sooner. On AMD's largest production design ever, they were able to get a 10x improvement, again, they were able to get the design spun in a single day. Very dramatic. Socionext is a formation of Fujitsu and Panasonic in Japan, they've had experience on multiple production designs over the last year and a half across many process nodes. Again, anywhere from 3 to 14x speedups. Samsung, also using IC Compiler II in production. They got similar kinds of speedups, in addition to that, very importantly, they got better quality of results.
They were able to reduce the error on their chip by about 11%, which translates to direct cost savings for them. These kinds of results have helped make IC Compiler II the fastest ramp in the company's history. We've started last year with four designs from some pretty compelling customers when we launched. We're now up to over 100 designs that we're tracking. 100 production designs. Many of them are FinFET, 16, 14, and even 10 nm, but many of those are also at 28 nm and 40 as well. At this point, we feel very confident about where we stand. More precisely, it's 117 production designs. 29 of those have taped out, and that's across 42 unique active customers and 18 process nodes.
How many customers is that?
42.
Okay.
This is a very impressive start over 18 months, but we also feel that we're at the early stage of what will be a multi-year upgrade cycle, as we have a number of our customers move over time to IC Compiler II and take advantage of it. IC Compiler II is clearly delivering some game-changing results for our customers, and it's really advancing our leadership position in digital design. Verification, as Aart mentioned, is also one of the biggest challenges our customers face. A year ago, actually a year ago next week, we launched our Verification Continuum, and it's targeted to help solve some of those problems. The problems are really driven by the exploding complexity of what's going on to a chip these days.
It's very common to see chips of 200 million gates plus, but what's more complex is making sure that all of this functionality, different IP blocks, all works together. Really, as Aart said, we're at the intersection of hardware and software, where there's a million lines of code on a chip, and people want to be able to validate that software works on the hardware before they tape out. Hardware-software verification is becoming increasingly critical, as well as just managing the sheer complexity. To address that, we launched our Verification Continuum, and at the heart of it is our industry-leading simulation tool called VCS. VCS is used by about 80% of the world's leading design teams and on the leading designs. That's important because simulation is still the centerpiece of a verification platform.
We've been working on multi-year deep collaborations with key customers to integrate all the hardware and software they need into a unified platform. Other people are talking about verification platforms, too. There are three key things that differentiate Synopsys. One of them is the fastest engines. Performance is absolutely critical in verification of any type, and we have the fastest engines. In many key segments, we also have the leading market position, the number one position. The second piece of it is unified compile across the platform based on the industry-leading compile technology from VCS. The third is unified debug across the platform based on our leading debug technology from Verdi. Again, we've made a lot of progress over the last year. I'll focus on just three updates today.
The first is we closed the acquisition of Atrenta in August, and that brings us the number one solution for static and formal analysis. We were just talking at my table about how important is that going to become. That's becoming a must-have now for verification, as people are able to find bugs earlier before they get to simulation. We feel very excited about Atrenta joining and about integrating that technology into the platform. The second really exciting area is emulation. As we've talked about on our earnings calls, we're seeing good growth in our emulation business. In fact, we see the whole emulation market as being very strong, benefiting all of us. We've made some key technology advancements over the past year. Most notably is bringing unified compile to our ZeBu emulator, and that's been critical for us.
Again, we've had a number of customers that shared their success at Design Automation Conference. I'll focus on two here today. One was AMD, who saw a 200x improvement in their verification flow by using ZeBu together with their virtual prototyping solution. That really transformed the way that they look at hardware-software verification. At Freescale, they were able to get 5 MHz performance off of ZeBu, and that allowed them to boot Linux in a matter of minutes, which was something they just couldn't do before, 4- 5x over what they had with their previous flow. The third area that we've made some great progress in is physical prototyping or FPGA-based prototyping. There, we've had the number one product for many years with our HAPS solution. Just yesterday, we announced our next-generation system called HAPS-80.
The hardware in HAPS-80 is based off of Xilinx's new Virtex UltraScale FPGAs, that together with our ProtoCompiler software, is able to let customers achieve 100 MHz performance, which really allows them to start not only verifying hardware and software earlier, but actually doing early software development. The other key enhancement is at 1.6 billion gates. Sometimes prototyping has been focused on just being able to deal with IP blocks. With 1.6 billion gate capacity, people can prototype their entire chip. A lot of progress with our Verification Continuum, a lot of progress with IC Compiler II, a lot of progress with bringing our IP into optimized solutions for IoT and automotive. I'd now like to turn it over to Andreas to give you an update on our software business.
Thank you, Dave, and thanks for having me here. I will talk about the Software Integrity business. I'm going to start to talk a little bit about the concerns of the software industry, then dive in some of the mega trends that we are seeing in the software industry, dive in a little bit into security, where we see us as Synopsys being a solution provider, talk a bit of our technology roadmap over the last year, and then go into customer execution. First of all, what I would like to do is go into the shifting concerns of the software industry. If you just look over the last 30 years, in the 1980s, with the upcoming of the PCs, software was really mostly concerned about single applications. We may have a Word processor or Excel spreadsheet on our computer.
It was used on a standalone computer at that time. In the 1990s, the internet came, and we started actually connecting all the computers, and we started transacting over the internet. We were doing bookings of hotel rooms, of flights over the internet. We were doing transactions of money over the internet and banking over the internet. In 2010, others calling it a smart of everything or IoT. I want to really focus on software-controlled devices. Software-controlled devices means, for example, in medical domain, my wife had a surgery a couple of months ago, minor surgery, and I was in the prep room, and every device in that little prep room had an Ethernet connector, was essentially connected to the internet. Whether it's infusion pumps or any measurement, everything is now connected. We talked about automotive.
Cars today have a lot of software in it and control a lot of the car really via software. If you look at this now from the concerns over time, in the beginning, it was really mostly about quality, but at that time for the application, it was actually sufficient that we got it almost right. If the application crashed once in a while, well, we weren't pleased about this, but we didn't go back to the vendor and would get a refund for our software. The actual implication in terms of the liability was fairly limited. This really changed when the computers got connected, and quality at the time was really supplemented by security, and the impact was significantly larger. Aart was talking about one of the large companies where the CEO actually lost his job because there was some security concern.
The issue here is getting it almost right or getting it mostly right is not sufficient enough because an intruder just needs one backdoor to hack into the system, and that one backdoor, which sometimes could be a quality issue, but that one backdoor is sufficient to essentially hack your system. Now when we go to the software-controlled devices, Aart mentioned, you also on the news, the recall of the Jeep, which was demonstrated to be hacked over the internet. That suddenly you have human health, you have human life at stake. Now the automotive industry, which really is primarily concerned about safety, because safety is really something where liability can be unbounded, pretty much. Safety, you cannot have safety in the car without security in the software, without quality in the software.
The way we see this is really quality as well as security, as well as safety, and increasingly privacy are all connected, are just different sides of the same coin. I want to talk a little bit about some of the mega trends that we see in the software industry. The first one is software seems to be always growing, and in fact, that's one of the reasons it's actually more slower that Aart was mentioning before. Look, it doesn't cost anything to increase the size of the software because memory gets cheaper and cheaper over time. Here are just a few examples. The Apollo 11 had, at the time, 140,000 lines of code. Today, we wouldn't be able to write any code with 145,000 lines of code that has any critical functionality.
If you look at Windows, from Windows 3.1 to Windows 7, the code increased from 2.3 million lines of code to 40 million lines of code. I wouldn't really say what I observed in terms of the functionality in Windows that the functionality increased by a factor of 20, but certainly the code size increased by that factor. Automotive, GM started in 1981, putting some software in with 50,000 lines of code. Today, you have 100 million lines of code . Just as a reference, and Aart mentioned that before, all of Synopsys code, all of Synopsys products is 400 million lines of code. That's a lot of code. That's really, it's a lot of code in there. 100 million lines of code in a car is pretty fundamental.
The second mega trend that we are seeing is really the number of software developers in the world. Depending on what source you're quoting here, some sources are saying there are 11 million- 12 million professional software developers. This particular source is saying 20 million today growing to 25 million in 2020. The number of software developers is actually rapidly growing. In fact, you can't even hire. If you try to hire a software developer in the San Francisco area today, it's extremely hard to get, for example, someone for a Java web application. It's extremely hard, very competitive market. Software developers are increasing, and software developers need tools. Software developers need essentially methodologies, tools, and training to get the software designs right. There's a tremendous need out there.
The third trend that we are seeing, just on the tools market, if you look here, the tools market for quality and security, this is actually fairly rapidly growing. Today we have about $1.5 billion in the quality space and about $900 million in the security space. Interestingly, security is growing at a 30% rate, much faster than the quality space. That has really to do with every time you see something on The Wall Street Journal on the front page, people wake up and saying, "Well, I need to do something about it." Chrysler is certainly highly alerted right now to security concerns in their domain.
The last trend I would like to mention, this is really an eye chart, it's really not meant to be readable, we see an increasing number of regulations, standards, even laws coming up that are regulating that space in terms of compliance. What you see here is really a maturing of what we call the software supply chain, where not only the software development process itself still has standards for compliance, but even the interfaces. When you want to ship software from a supplier, from a tier 1, for example, to an OEM or from a tier 2 to a tier 1, there will increasingly standards that are driving the quality and the security of the software that's being shipped. Dave was mentioning ISO 26262 in the automotive domain for safety criticality. There's actually part of that that's applicable for software.
This is part of that that's applicable for software. This is an increasing domain where we feel there's an opportunity in terms of playing in the compliance space, providing solutions in terms of tools, methodologies, and training, is really our way to go on the business side. Let me talk a little bit about security and how security really evolved over time. In order to do this, I would like to separate between development, deployment, and production. Development means you have a development team that gets the requirements in from the marketing department or wherever the product requirements are coming from, doing the coding, do the testing, all the way till it gets to the release.
Deployment means you're now shipping the software, you're getting it out on the target machine, you're installing it, you're configuring it, you're ready to go, you can flip the switch. Production meaning you actually flip the switch, your website may be online, your database is online, and you're actually using it on a daily basis. If you look at this from security solution, it's actually very interesting that security grew really from the right to the left. Security solutions really started originally with firewalls in the early 1990s. Firewall is essentially a filter on the periphery of your enterprise that is looking at the traffic and filtering elements of the traffic out that are not needed. In the beginning, it was just on the network level. Then now the more intelligent firewalls are really growing up into the application layer.
This is a market that's about $12 billion, that's growing at 7% year-over-year. What's interesting about the firewall, they work actually quite nice. However, they're always an afterthought. In a firewall, you filter out attack vectors that you know, but the hackers are very good at finding always new attack vectors. In that, it's always an afterthought. In that sense, it's actually leaky technology. Over time, security really moved further to the left and what's called penetration testing. I'm sure all of you, working in your institution, you have consulting firms coming in that do a penetration testing of your IT infrastructure. Penetration testing is nothing different than you try to hire some hackers and saying, "Try to hack me." Right? The hackers coming in maybe for two weeks, maybe for three weeks, is trying to find backdoors.
In the same way that a malicious attacker would try to find a backdoor. You hire him, you do this for a few weeks, and you find the most vulnerable elements of your IT infrastructure. The problem here, it's not scalable. It's not scalable in terms of you're never done really with penetration testing. You could just do it as an ongoing process. Also, security experts are even more difficult to hire than Java developers, for example. It's a very, very rare talent having experience as a security expert. This is just not a scalable solution. What you really see increasingly, that security is really complemented now by moving to the root cause of software development, meaning, doing software development in a way that you avoid a whole number of vulnerabilities building into the software in the first place.
Meaning, security and quality, and now increasingly safety as well as privacy, built into the process itself. This is where you see names like HP and IBM playing Veracode, for example, and this is where we as Synopsys are really increasingly playing. In summary, where we see us really is on the development side of the software lifecycle, addressing quality, security, safety, and privacy, which, what I mentioned before, are different sides of the same coin, and building those into the development process itself. Let me talk a little bit about how we came about in the last 1.5 years. As you all know, Synopsys started the software business by acquiring Coverity in February 2014. We have expanded since then in 2 dimensions. The first dimension is organic. We added organic investment, and I want to just mention 2 areas here.
One is adding additional programming language support. This is really addressing the fact in the modern enterprise today, they use probably a dozen different programming languages. In order to be a comprehensive solution provider, you really need to have solutions for all of them. We started that investment earlier this year, already we're having more and more languages coming out of our pipeline. The second area is compliance. I mentioned that earlier. Essentially addressing the different compliance standards, and in fact, it's also driving some of the compliance standards in the future, is a key investment area forward going. The second dimension I would like to mention is really non-organic investment, and I would like to talk about 2 acquisitions that we have done in the last half year. The first one is Codenomicon. Codenomicon is a well-known company in Finland.
They were the co-inventor or the co-finder of the Heartbleed vulnerability that many of you know, that was the SSL library vulnerability, which suddenly the entire internet was pretty much vulnerable to that vulnerability that was found. With Codenomicon, we acquired a world-leading fuzzing technology. Fuzzing technology is really finding vulnerabilities on protocols. Dave talked about earlier the Ethernet in the car, for example, the CAN bus in the car is a protocol. Or for example, on your device, the Bluetooth is a particular protocol. One of the fuzzing technologies, for example, is saying, "Let's take a Bluetooth device, let's fuzz it and try to find essentially a vulnerability by exploiting some of the protocol side effects and hacking your Bluetooth device on that." We are the world lead on that with 250 protocols rapidly growing, addressing multiple verticals in this domain.
The second technology is software composition analysis. Software composition analysis, this helps actually for existing software, scanning that software and finding known vulnerabilities that, for example, come in open source libraries. Very often when open source is released, later on, security researchers find vulnerabilities and the software is now distributed and present in many components. This product allows us to find those, mark it, and then alert the customer, "You should actually upgrade the corresponding release." The second acquisition I would like to talk about is the acquisition of the Seeker product from the company Quotium. This actually expands our solution into dynamic application security testing. This is a new technology area, what's called an interactive application security testing. We are a world leader in terms of that technology. This is really helping us now expanding into web application and web application security testing.
Let me talk a little bit about our field execution. Our sales channel is really composed of two major pillars. The first pillar is our existing sales channel in Synopsys, selling to the existing Synopsys customers. Aart has mentioned before, our customers have actually for multiple years already, more and more invested into software and asked us all the years, "What can you help us on to help in the software domain?" The emulation technology, the prototyping technology is all helpful. With our Software Integrity solutions now, we're expanding that for the existing customers. The second one is a dedicated enterprise and sales team that is really focused on selling to enterprise customers that are outside of the existing customer base. For example, New York Stock Exchange or Nasdaq. These are customers that are new to Synopsys, and the dedicated sales teams will really handle those.
Both of them are really built on two foundations. The first one is a deep culture of close customer collaboration. This is something that Coverity has in the culture, Synopsys has a very long culture on that. In order to drive success for customers, it's not just about giving them tools. It's essentially transforming them in terms of the maturity, how they do software development. That's a deep culture that Synopsys brings to the table, collaborating with the customers to drive them forward in the maturity. The second one is a strong international presence. We have 113 sales offices in 27 countries. This is certainly a very broad infrastructure that we have from which we can operate and essentially serve our worldwide customers in a fairly rapid way. Let me share some of the customers that we have here in the different verticals.
You see on the top left, mobile consumer devices. On the right, in enterprise networking, in independent software vendors. Bottom left, high reliability devices, for example, in the medical domain or automotive domain. Then on the bottom right, you see financial services. What's really interesting, the technology stack that we have, the technology solution and the products are actually equally applicable to all of them. The concerns are sometimes different. Just to give an example in the financial sector, a bank cares actually about two things. Number 1 is it cares about security, that the bank account doesn't get hacked. If the bank or if a financial institution has a trading platform, the trading platform cares about quality. Because a trading platform going down, every millisecond our trading platform is down. You guys know this very well, right? It costs you millions of dollars.
Quality and reliability is extremely important there. Let me talk a little bit about going back to supply chain. One of the big opportunities we see is providing solutions for compliance in the software supply chain, meaning software is not just shipped based on the fact that the code compiles. It actually complies with certain criteria for quality, for security, for safety. This is an area where the automotive industry, for example, supply chain, and we had that discussion at the table just earlier, is very advanced in supply chain. In software, this is really an emerging area where we see a tremendous opportunity. I would like to quote our collaboration with Underwriters Laboratories. I am sure all of you know UL. An average household in the U.S. has more than 100 devices that are UL certified. UL is about to launch a cybersecurity assurance program.
We are collaborating with UL, providing our solutions, but of course, UL is also utilizing solutions from other vendors. This is really an emergence of compliance in a market that is not used to that, and this is a tremendous opportunity for us. Let me talk a little bit about some of the coverage that we received recently. One of them is Gartner. Some of you may know the Magic Quadrant for application security testing in Gartner. We just entered their Magic Quadrant on the visionary side, and that is a tremendous accomplishment for us. A lot of enterprise customers look in Gartner for guidance in terms of which vendors they should utilize for the different technologies, and having Synopsys there now as an emerging player, I think is really great opportunity for us. The second thing I would like to mention is Forrester.
Forrester in their TechRadar for Application Security, we are actually present on four of the key technologies. If you look at there, I just want to mention this here, very little print, and you see it on your handout. We have application firewall. This is what I mentioned earlier on my chart. This is really the firewall, meaning, at the production side of the software. We have really the technologies to address security in the development process much earlier. This concludes my part, and I would like to thank you very much.
With that, maybe we can open it to Q&A for any of the three of us.
If you could hang on-
Oh, I guess there's a microphone that's going to be passed along.
That would be helpful. How did I know this was going to be here?
Good afternoon, [Geus and Lisa . This is] Howard. Question for Aart. The implications of smart products, of which IoT seems to be a special case, in terms of how EDA tools and IP are employed. It's basically a two-part question. One, are there significant differences in the requirements of smart connected products versus prior generations of new devices over the last 30 years? Secondly, we've seen from many customers, particularly in automotive, a move towards working on common platforms or products around which there are significant variances or variations. What are the implications of that for how EDA and IP are employed? It's pretty easy to see how it affects your CAD and TLM and other kind of technical software counterparts, but what does that mean in terms of design methodology for EDA?
Excellent question. I think the first aspect to answer that question is that these products, because they have not only very sophisticated silicon, but unbelievably sophisticated software sitting on top of that, the first question that a developer company has to answer is, does it actually work? Does it actually do the thing I want it to do? Long before actually building it or building it into a car, let's say, or another very expensive device. The way to answer that question is what's called prototyping. Do a mock-up. In the past, we would have said a simulation. Now, simulation is not fast enough to do that. Simulation amended with other techniques such as emulation or FPGA's case. One of the areas that we see tremendous success in the Verification Continuum is that it's actually a continuum from hardware to software.
That has accelerated substantially because when we quote things such as, oh, you can get Linux booted in an hour or Android in an hour, what does that really mean? That means that this mock-up within an hour can tell you, does it sort of run? Within an hour, you can start running applications on it, which you can try it out. I think that is a massive direction that we will see more growth in and more opportunities in. That also implies having the sophistication to do what is very large systems. On the platform side, why do people use platforms? Well, because they hope to use the same thing for different applications by modifying them a little bit, A. B, a platform invariably means multiple things, whatever they are. A car is a platform, a computer system is a platform.
Our software is two or three platforms, where the value of the individual pieces when they work well together in a platform is much higher than the individual pieces in separation. When you ask the question about IP, one of the reasons we sell not only individual IP pieces, but for example, we have a sensor subsystem, is because in a sensor subsystem, there's processing, there's memories, there's interfaces and so on. If we put it together and let's say you want to connect your sensors to it, well, at least that part is already working. Yes, platform is just another way of saying put more things together and the next level platform is higher. Now back to the first part of your question. We always talked about the verification platform as a silicon thing.
Now we don't think about that at all like that. We think as silicon and software thing. Well, we did exactly the same. We expanded the scope of what needs to be solved. The sophistication of that solution is very similar to what we've seen in 20 years of Moore's Law sophistication now applied straight to the software domain. Yes, please.
Hi, Aart. I guess this is a follow-up on that question around smart devices. As you build up your blocks and You have this IP solution that you could start offering. When I think about customers for smart devices, it's expansive, right? It's not the typical customers that are buying-
Yes
typical semiconductors. In the early discussions that you were having, where are you seeing the most interest? Is it a typical semiconductor company? There are many semiconductor companies that are working on these fundamental blocks, whether they're processor sensors or what have you. Or are you starting to see system OEMs come and say, "Wow, we could take your IP and then add our value add and come up with our devices." Where is the interest, and if it is the system OEMs, are we going to see an expansion of TAM because of that?
The answer is yes.
Okay.
Meaning we absolutely see both, of course, there's been already a long-standing push on the semiconductor side of things because you may recall in the late 1990s there was this new word, system on a chip, and it sounded big, and today it's like, huh, okay, duh. Everybody has systems on a chip because most sophisticated chips are one or more processors in the first place, have all the characteristics of being little super small computers. The semiconductor industry is well-versed in knowing that on these little computers you need some operating system, you need the drivers, and then you need the way to connect above. The challenge for the semiconductor industry, by the way, is not so much a technical one, but an economic one, which is, how do we get paid for that?
They are racing forward in trying to provide platforms. On the other hand, there is very clearly now a perspective coming down from the people that say data is intelligence, data is insight, data can impact devices. Again, I think an extreme case, extremely interesting is actually the automotive industry because it is suddenly very swift that we see this driving automatically is not only possible now, the question will it be legal and what are the side effects of all of that? Those are people that come down from the system side, and they have now the opportunity, the dilemma, the question, should they do the chip design themselves, and if they see opportunities there or if they have special requirements, such as automotive has certain manufacturing requirements.
A Bosch, for example, will say, "Hey, we can do a lot of that ourselves." Others may say, "Well, who do I team up with? Do I team up with an NXP or a Freescale or an ST that are the big automotive chip providers?" These value chains continually realign themselves essentially based on how can you be more competitive, i.e., how can you move faster? It is an interesting, difficult question in the automotive industry because the automotive industry is fundamentally a slow-moving industry. Now the cars have to work for 25 years. There is all kinds of requirements, and suddenly it is turned into a super fast industry. This is artificial intelligence in real time here. It is hard to predict all of these things, but you can feel the next wave, in my opinion, is absolutely that computation, mobility, and so on, gets aggregated into distributive smart.
Until it is there, the hunting for money will be challenging because it is hard to make money on some temperature checker, right? The car is a lot of money. Super interesting direction. I think we are well-placed for that, and we ourselves are emphasizing now more vertical market segments than we ever did before because we are seeking exactly the characteristics you are asking about.
Aart and Dave, maybe a follow-on to that, taking it into your IP business. You talked about the IoT and the auto platform. The sense that we have always had is that the margins in the IP business are lower than the corporate average, and again, I think the sense that many of us have is that it is the extra work that you have to do with each of the customers. Now that you are coming out with these platforms that are designated to a vertical, are you finding that there is more standardization, meaning that the customer can take the IP, for lack of a better term, as is, so you get more repeatability and maybe more leverage out of the IP business?
Want to take that, Dave?
Am I on?
Yeah.
I think the answer is yes. I think we'll see that. On the counter side of it, the challenging part too is when these standards evolve, when new versions come out, they become much more complex. The initial R&D investment that needs to go in, for example, to go from USB 3.0 to 3.1 is huge. It's like a 10X in complexity. I think that will also drive our customers to want to take things off the shelf versus mess around with them. One of the key terms we use a lot in the IP business is silicon-proven. The fact that our IP has actually been on a real test chip, and we've been able to test it, has huge value to the customer. By having it out of the box, they get the advantage of that.
I think these moves towards these platforms will help encourage customers to take more off-the-shelf IP.
Yeah, they're already taking it. When they take.
The sense is that you're still putting a lot of manual design effort after they take it in, which kind of weighs on margin. Is there more opportunity to just take it as is?
I think actually the bigger margin pressure is not from that angle. It is from the fact that right now the rate of change on silicon technology is unbelievably fast and complex. I'd like to remind you that seven, eight years ago, the notion of there ever will be commercial FinFET was no. Here it is, you can see the rate of change. That is an opportunity for us because that complexity is also very difficult to handle by the users. Now tying to the fact that they will need IP that fits their restrictions is absolutely key. Recently we introduced the fact that we have a whole IP collection that is automotive certified.
With other words, we did all the homework for that market segment, therefore they're more likely to come to us for that IP because it passes what are actually very strict checks. I think all of these markets right now are in fairly fast development. For us, it is absolutely essential to figure out what can we horizontally leverage and what things do we need to invest in vertically. Invest is, of course, expensive, there's also specific benefits and pricing that you can leverage by going vertically. I think the IP business is in just as much evolution as the hardware-software interface because IP is hardware-software these days. Yes? Oh.
Thank you. When you guys acquired Coverity, if memory serves, you talked about a TAM of around, I think, $500 million at that time for what Coverity did at that time. You had a TAM for this year now between security and quality of I think around $2.4 billion or something of that magnitude. Can you talk about how that TAM has evolved, and do you really serve that whole TAM today with the pieces you've acquired and organically developed? Thanks.
At the time, I think the TAM that was quoted there was really the quality space which we played at the time. Remember, we were in a static analysis quality space, although we sold to security. Now when they quote the north of $2 billion, that is really the combined quality and security as well as static solutions and dynamic solutions. The full testing spectrum that you're talking about here.
Thanks. Could you tell us where you are seeing the fastest growth by specific nodes, regions, and customer types?
Okay.
Just a little bit color.
Sure. Let me start with regions. The Far East is still the most rapidly growing, and notwithstanding sort of the recent up and down and shakeups in China on their respective Wall Street, the investments continue at a very high rate, a large number of engineers coming out of school, and definitely a drive by the country to accentuate its growth rate in the semiconductor areas.
Which country?
In the semiconductor areas.
Which country?
I'm talking about China, but in general terms, I'm talking about Asia-Pacific. Asia-Pacific, I would certainly include Taiwan and Korea and China as the big three. Then around that, you would look at the Singapore, Hong Kong, and somewhat India space. That's for us, Asia-Pacific. We take Japan in a different category. That's the highest growth from that perspective. In terms of the growth rate at the different technology areas, when we look at EDA, we typically say we're talking about low to mid-single-digit growth rate. About the IP, we're talking about just double digits. Then the software space, 20% plus with a lot of variability as you see these markets being very fractured and evolving.
Then from a technology point of view, I personally think that the verification space is one that has a lot of potential because the problem set is becoming so much larger and more complex per what we talked about and the intersection of hardware, software, certainly a center of gravity for us.
Could you talk about the FinFET activities that you're seeing at Foundries versus IDM and
The what?
The FinFET activities.
Sure.
By nodes, that would be helpful.
Sure.
By nodes, that would be helpful.
First, just to clarify for those not familiar, what is FinFET? Transistors for the last 40 years were flat and smaller and smaller and smaller and smaller. Then it's hard to make smaller, so what do you do? They flipped them vertical. It had not only the benefit of using less real estate, it had another benefit, which is it was also a breakthrough in getting lower power. Low power, as you clearly know from your cell phone and other mobile devices, is actually the single biggest physical challenge. That is what FinFET was addressing, and the question was, can you manufacture these at a reasonable cost? It's actually really difficult, right? We have horizontal houses is one thing.
When you build high-rises, it's a whole different set of skills, it's not dissimilar from that perspective. The answer has been yes, it is possible. It has also been clear that it has been very difficult to get there. One of the reasons we see so many generations of FinFET in a very short amount of time is because people keep improving them so that the yield, meaning the number of chips that work over the overall number produced, needs to come up because that has a direct impact on the economics. The race has been very much on among the top leaders from a technology point of view, and that is still ongoing because the horizon keeps moving out. What was viewed as almost impossible, 10 nm chips, is now in production design.
Now already people say, "Well, we want to go to seven, we may go to five. There may be other types of transistors possible." I expect that to continue. I expect the economics to become tougher, because of the value of smarts at a low power cost will increase. There's no doubt in my mind that the push will continue. Not everybody is adapting at the same time. There is a large set of companies that are sort of in the wait-and-see mode, and they are essentially parked at 28 nm, maybe 22 nm, and they're just waiting until FinFET is cheap enough, easy enough to then cross that bridge.
You may have noticed that in our IC Compiler results, we made a point to say, here are a number of chips that actually were done at 40 and 28 nm because the many people that benefit from these advanced tools, even if they're not on the singular most advanced silicon technologies. Bottom line from my perspective is the technology of Moore's Law is continuing at a rapid speed. The economics of Moore's Law are under some degree of pressure, the functionality value, in my opinion, will be so high that it will push the whole system to keep racing forward.
We have time for one more question.
We've heard a lot from semiconductor companies at recent conferences kind of updating their views on the macro outlook. Just kind of curious, as we're getting ready for most of your customers to go through the budgeting cycle, what's some of the commentary that they're coming back to you around their thoughts around investment in EDA? Maybe if you could layer on top of that, given the wave of consolidation, which you talked about on your call cadence as well. Have you actually gone back and looked at specifically your customer set and the consolidation and when we might see, if any, the impacts that we've talked about? When does it layer in from a timing perspective?
Sure. Well, there's no question that specifically also with some of the recent events in China, there's a bit of reluctance to see super high growth in semiconductor for this year and potentially for next year. People are revisiting their budgets. Secondly, you have seen a number of consolidations that have been quite substantial, at least announced. Not all have been closed. NXP-Freescale, Avago-Broadcom, and so on. This is not new. These are quite large, and I think they're all part of what I love to call the techonomic crunch of continuously redriving for efficiency. Now, when you say efficiency, you say, well, they're going to come to you and say, "Oh, I want to pay less." They always say that. There's nothing new. We've always managed over time to still grow the company. In consolidation, what happens?
Well, initially, the customers say, "Hey, I need to save money because I paid a premium for the acquisition. I'm under pressure." Then we listen, then we say, "Okay, let us help you save money by being more efficient, by maybe doing more things with us." In most cases, we've actually been able to navigate well through this. These are the normal churns in the industry. I think the bigger picture is one where many of these companies struggle with the very question that we're addressing, which is that a substantial amount of their value is delivered through the software that's embedded in their chips, and that they are seeking to find where are the new opportunities that actually will have substantial volume.
Our best move here is to help them be competitive in being able to prototype very quickly what are the products that they can offer up. I've always believed that as much as it's important to help our customers save money and be super efficient, it is even more important to help our customers differentiate and help them grow their revenue. Trying to be as well aligned as possible with who we believe are the winners and trying to provide solutions that will make them competitive is the core of our answer, and hopefully, you saw it a little bit, the core of our strategy. Okay, I'm getting this sign, which implies universally it's sort of over. Thank you so much for joining us. Hopefully, it gave you a sort of comprehensive and coherent sense of what we're doing.