Judy Warner (00:01.123) Hi, Simon. Thanks so much for joining us today on the podcast. I look forward to talking to you about all the exciting work you're doing. Simon Rance (00:08.684) Hi Judy, how are doing? Judy Warner (00:11.033) Good. Well, for our audience, you have an interesting background in the sort of EDA space and tell us a little bit about your background and sort of your technical interest and what you do for Keysight. Simon Rance (00:26.702) Sure. So I have quite a broad background. I've been in EDA for about 25 years or so through all aspects of it from marketing, sales, product management, application engineering. So kind of a good breadth of that across digital analog mix signal and even some RF microwave as well. Currently, my role at Keysight is I'm director of product management for what we call our process and data management. portfolio or business unit. And we have two products that I currently manage. One is our Design Data Management Solution, SOS, and our Engineering Lifecycle Management Solution, HUB. And these two tools and myself joined Keysight as part of the acquisition of CleoSoft about a year and a half ago. Judy Warner (01:19.848) Okay. So tell us a little bit about both those tools and then let's talk about from an EDA standpoint. I have some following questions. So tell us a little bit about those two products and then I've got a follow -up question for you. Simon Rance (01:38.862) Yeah, sure. So we have two products, I mentioned, CleoSoft before Keysight acquired the company had been around for about 26 years. And so we have quite mature products. Our most mature product, which is what CleoSoft was originally founded upon, is the design data management product that is geared towards hardware silicon and semiconductor design. Judy Warner (02:05.603) Mm -hmm. Mm -hmm. Simon Rance (02:06.22) And the other product, the Engineering Lifecycle Management Hub product helps with IP management, traceability, essentially addressing kind of a PLM or PLM Lite type solution that is dedicated for an engineering lifecycle for semiconductors and silicon chip designs. Judy Warner (02:25.463) Okay. So since I sort of come from the hard, the PCB board level side of the fence, how does the data management, how is that handled differently between semiconductor level versus say software or like PCB data management? Simon Rance (02:48.814) Yeah. So one of the main differences is that for hardware engineers, there's a lot of data compared to, let's say software design. So there's a lot of data management solutions out there. Primarily they are geared towards and focused on software development and software development teams. Whereas we saw a niche or a capability that was missing in the hardware design world about 20 plus years ago where Judy Warner (02:57.961) Mm -hmm. Simon Rance (03:17.994) managing large design files like schematics, layouts, especially for analog design, for example, and mixed signal. And then how do you manage that across multi site teams, either within the same region, country, or even globally as well? So these are some of the unique things that started to make us realize that we needed something more geared towards hardware development, specifically for chip design, that the software data management solutions were providing. Judy Warner (03:24.153) Mm Judy Warner (03:34.884) Mm Judy Warner (03:49.369) So tell us a little bit about, so you found this niche, right? And so I can imagine semiconductor designs are extremely dense. so what are, does this data then, do those products, since it's an EDA, you know, data management PLM tool, did those live in the cloud or the on -premises? Are there security concerns there? Simon Rance (04:14.286) That's a good question. have customers that use our solution on premises and that's the privacy, the primary or the most common scenario. We have some that do use it in the cloud as well, but typically with the cloud, it's a hybrid model where parts or most of the data is on premises. And then when certain things need to happen, let's say if it's you're running some verification that you need to really crunch down that time to verify. Judy Warner (04:24.612) Okay. Judy Warner (04:32.259) Mm. Simon Rance (04:44.706) A lot of the time that data then gets moved to the cloud to get that processing power and the compute power necessary. And that's where we have those capabilities and we have customers that use our solution for that hybrid on -premises cloud type scenarios. Judy Warner (05:00.129) Okay, makes sense. So one thing that I've been observing while dealing with various people across Keysight, but across the industry that I'm really encouraged by is the ability that people are starting to collaborate, right, more digitally and giving the engineers a better experience to sort of stay in a limited number of EDA tools even though they might be different tools. Do you offer something like that where it can work alongside different EDA tools? Simon Rance (05:39.116) Yes, that's one of the key differences between data management for software versus hardware design. With EDA design flows, because of the complexity of those flows, the methodology and the tools that are used, our customers are having to use very complex tools. It builds a lot of data very quickly. So those file sizes and the sheer amount of data grows very quickly. Judy Warner (05:47.97) Okay. Judy Warner (06:00.899) Mm -hmm. Simon Rance (06:08.488) One of the big things is that design engineers and design hardware design engineering teams, they want to minimize the chance of error, one of the things, as well as security concerns, right? So the more that data gets moved in and out of multiple tools, the more room for error, just because it's a natural thing that will happen, right? As you're moving and importing and exporting data in and out, but it can also then... Judy Warner (06:23.544) Right. Judy Warner (06:28.857) Mm, mm -hmm. Simon Rance (06:36.684) what happens if you don't import or export all of the data right into the next tool? You can have other processes and flows downstream that you don't realize became a catalyst from that point of view, right? this one to two weeks later, they realized, we took the data out of this one tool, put it into this other tool, but didn't have all of the information it needed. so verification run was incomplete or our test results are incomplete. So what, well, Judy Warner (06:57.614) Mm. Simon Rance (07:05.94) What our customers have been asking and looking for for quite some time now is having the ability to move the data less between tools and flows and have them native side of their EDA tools. Judy Warner (07:15.757) Hmm. Judy Warner (07:19.151) Do you think that's a market driver for what I'm saving? Simon Rance (07:23.744) Yes, definitely. We have our solution as a standalone, but most of our customers use our data management solution natively inside their EVA tools so that they don't even have to leave them. If they use more than one tool, they can use the data management in both of those tools and not have to move the data between them. Judy Warner (07:25.526) Okay. Judy Warner (07:45.401) So how do your tools connect with other EDA tools? Like what is that connector or what is that experience like for engineers that may be in our audience? What is that like? Simon Rance (07:57.112) Yeah. So we have tight integration with all the major EDA tool partners, Keysight. One of them is our Keysight ADS for RF and microwave. That user experience is if you're inside of Keysight ADS, you have just there in the native menu system, you select our data management solution. You can check in, check out files, you can tag, you can do branching, you can do everything you can do in our standalone. Judy Warner (08:17.826) Mm Simon Rance (08:27.096) data management solution, but you can do it natively within that tool. And if you have another tool in the flow that may be from one of the other big vendors like Cadence or Synopsys or Siemens EDA, there are tools as well. have them, we have plugins and integrations natively inside of those so that our design engineers are using both solutions, ours and those EDA tools can securely manage data. Judy Warner (08:49.678) Okay. So if they're in, Cadence or inside Mentor Siemens EDA, then they can stay in their design tools and still access that data. Simon Rance (09:05.46) Exactly. So if it's analog design, they may be using Cade's Virtuoso. They can just access the data management from their menu system. Judy Warner (09:10.52) Right. Judy Warner (09:14.807) I really love this trend. think it's making things easier, I'm sure. So when we're talking about this, you're definitely talking about enterprise teams. take it. Simon Rance (09:25.398) Yes, enterprise teams, a lot of collaboration, a lot of traceability that's needed. Judy Warner (09:32.987) So let's talk a little bit about that. So traceability, there's obviously some applications like automotive or military that have rough demands on traceability and security in that. what kind of applications do you see? And what is sort of ensuring that you get that traceability? functional and working in there so that, you know, we don't crash our cars or just. Simon Rance (10:07.49) Yeah, exactly. So that's where our other solution, our other product, the engineering lifecycle management solution comes into play. And that is, really, when you look at large organizations, or even medium sized organizations, different teams will use different data management solutions. And it may not be the same solution, because in the past or historically, they worked more in silos, right? So a verification team. Judy Warner (10:17.784) Okay. Simon Rance (10:34.872) doesn't always necessarily work directly with the hardware team or the software team or the test and validation type teams. So they'll pick their solutions, but you need to basically bring all of those data management systems together under one single source of truth. And that's where our engineering lifecycle management solution hub basically integrates natively into not only our data management solution, but all of the other major ones out there that are used for hardware, even software design as well. So that you get this holistic picture of the entire engineering life cycle. All the data, no matter where it actually may live, it could live on premises, it could be in the cloud. It doesn't really matter to our solution because we can manage both simultaneously. And so by having that single source of truth, then you can have true end -to -end traceability that's really the critical requirement for automotive applications that require functional safety. And ISO 262 Judy Warner (11:13.923) Mmm. Judy Warner (11:32.855) Right. huh. Yeah. Simon Rance (11:33.77) to the same with the military aerospace applications. Judy Warner (11:39.523) What are the years of traceability requirements these days? Like years wise, how long do have to hang on to that? Simon Rance (11:49.998) That's a really good question because I would say for a typical design cycle, a lot of the times it's just erased within weeks to months after the design cycle. But for industries like automotive or military aerospace, typically the requirement is a minimum of five years to 10 plus. It's really the life cycle of that device or system being in the marketplace for any type of audit trail and traceability. Judy Warner (11:58.095) Yeah. Yeah. Judy Warner (12:04.334) Mm -hmm. Judy Warner (12:10.136) Yeah. Judy Warner (12:17.326) Mm Simon Rance (12:19.896) But we're seeing some of those requirements that were exclusive more so to those industries are now coming into play. Even the semiconductor consumer, IoT, server markets as well. Because if something does go wrong, once that device or system goes into the field, if you don't have the data to trace it all back, you're pretty stuck. really for most semiconductor and Judy Warner (12:26.137) Mm -hmm. Judy Warner (12:42.574) Right. Yeah. Simon Rance (12:48.652) hardware system design companies, it's up to them on how long they typically save the data for. But more doing it just like the automotives and military aerospace industries and more starting to look into what types of data do we save and for how long for AI ML type applications for future designs. So that's kind of a new area of that as well. Judy Warner (12:54.051) Mm. Mm -hmm. Judy Warner (13:14.617) Back when I was in the PCB fabrication, working with large OEMs and Mill Arrow, of course ITAR was mandatory. And before digital, we had bonded rooms, right? Where we could keep data and all of that. How do you manage, is there a guideline for say ITAR? Simon Rance (13:26.062) Mm Judy Warner (13:41.795) that as you store these massive amounts of data and that you have all this available and maybe for years, like how do you manage things like ITAR? Simon Rance (13:54.19) through a lot of different mechanisms from, you know, access controls from tagging to branching to just preserving the information as well as securing the information as well. So that's a big aspect of it is the security aspect of IP who gets access and see what and that's that that's becoming a bigger thing. The more the more collaboration there is, the more chances security breach. So it's that fine tuning of Judy Warner (14:10.755) Mm Judy Warner (14:15.883) Mmm. Judy Warner (14:20.078) Right. Judy Warner (14:23.576) Yeah. Simon Rance (14:23.746) How do we enable more collaboration, but also increase the security of the IP or the data? But ITAR in itself has kind of its strict kind of guidelines that you have to follow on properties. But most companies even impose even more on top of that because, you know, just indemnification, legal type concerns. Judy Warner (14:34.348) Mm -hmm. Judy Warner (14:42.861) Right. Okay. Okay. And it is interesting in global economy, who can look at what? Right? It's a challenging thing to manage. You know, we didn't talk about this ahead, so I'm going to sort of put you on the spot on this one a little bit. do you have, can you share with our audience something like a really interesting case study? Like this is how their design flow went before and then they... Simon Rance (14:49.666) Yes. Judy Warner (15:12.887) integrated data management with this PLM and the rainbows came out and the unicorns or whatever that it was one way and it's another. Simon Rance (15:21.28) Yeah. Simon Rance (15:27.502) Yes, I can. Some of the really good ones we have are in kind of more the automotive military aerospace. So I'm not able to share exactly those particular case studies and success stories. But we have had one recently, which you'll be able to find information about that on keysight .com. And that's with one of our great customers called Allegro. And Allegro is used both are Judy Warner (15:36.93) Right. Judy Warner (15:54.583) Simon Rance (15:55.214) the West data management solution for hardware designs, but then also our engineering lifecycle management hub. And from managing that entire lifecycle, everything from the specifications, the requirements, the architectural spec, all the way through to tape out, all of that is managed from end to end, not just the data, but even the engineering know how. as engineers start to integrate IPs together, maybe an IP won't connect one to one to another IP, what type of glue logic do they have to create to get that to stitch together? That is actually being captured as well. So that intelligence, that know -how has been added to that IP, to the system so that on a future project, engineers can say, this is how these two IPs were actually successfully integrated together on a previous project. We can just leverage this glue logic that's also been checked into the lifecycle of that project. Judy Warner (16:26.851) Mm Judy Warner (16:47.044) Right. Simon Rance (16:54.126) You have all of those kind of going through documentation being managed, reports auto -generated so that you have that bill of materials and hierarchical traceability so you know what's being used by whom and where in the company, as well as any other type of restrictions that might be there security -wise, for example. But for Allegro, their case study was a 50 % improvement overall in that entire engineering lifecycle. in terms of productivity, time and market savings. Judy Warner (17:22.744) Wow. Judy Warner (17:26.201) That's huge. That's really huge. Sure took a minute to get it set up, but like how much time that will save over, it'll probably be millions or billions of dollars. It's, that's crazy. Simon Rance (17:27.981) Yes. Simon Rance (17:36.982) yeah. The more that gets preserved, kept, managed data wise, the more the savings are, the productivity boost, the time to market savings. And that does vary from customer to customer. The more time that that customer has to invest in putting more knowledge, more data into the system as they go along through the engineering lifecycle, the more they can make out of it in terms of reuse in the future. Judy Warner (18:00.578) Mm -hmm. Mm -hmm. Judy Warner (18:06.583) Yeah, totally makes sense. Sounds way more efficient and wise, especially with the load engineers have right now and the complexity. So, the way that I really was introduced to you, Simon, was I know not too long ago, you did a talk with Keysight's EDA World Connect Tour. Before I let you go, can you... Simon Rance (18:12.492) Yes. Simon Rance (18:17.23) Yeah. Judy Warner (18:34.071) Tell us a little bit about that and then I want to get that link from you. So if people are interested in learning about sort of innovative ways to handle their semiconductor data management PLM, they could go check out that talk. Simon Rance (18:50.262) Yeah. So what I did was I gave a bit of a high level executive summary of what exactly the type of problems that we're addressing with both the design data management solution, as well as hub for the engineering life cycle management and how the two together can give you the type of benefits that say Allegro was able to accomplish. And I kind of take it through one by one how that's done as a process at a high level. just to give a bit of insight in how that's done. But then I also touched on a lot of other type of common day -to -day engineering lifecycle problems that people will see or go through and how we address those as well. And so I can't go through every problem that our solution addresses, but it's interesting how many engineers just do the same thing over and over again, not realizing that Judy Warner (19:28.013) Mmm. Judy Warner (19:38.917) Right. Simon Rance (19:46.37) They can save a lot of productivity if they just solve it a little bit differently this way and with our solution. Judy Warner (19:49.315) Yeah. Yep. Yep. That's the challenge, isn't it? To stop doing the things the way you've always been doing them. To maybe pick up all that productivity and get out of your comfort zone. Well, Simon, thank you so much for that. And I'll make sure that I share that in the show notes and send folks that way. Cause I know we don't, I would love to podcast don't give you, nobody would listen to me. Simon Rance (19:54.648) never Simon Rance (20:00.0) Yes. Simon Rance (20:11.576) I can't sleep now. Judy Warner (20:18.881) Or maybe you either Simon talking about, you know, but I want to give folks a chance to go do that if that's an area they're interested in. So thank you again for coming to on the podcast and telling us a little bit about your talk and good luck with your, with your, with both your projects and hopefully in a little while you'll come back and update us as you get down the road a little further. Simon Rance (20:22.094) I'll read a little bit. Simon Rance (20:46.604) Yes, definitely. I appreciate it. Thanks so much, Judy, for your time. It's great to be on your podcast. Thank you. Judy Warner (20:50.691) Thank you. Yeah, thank you. For our audience, thanks so much for joining us today. I hope you enjoyed this sort of executive view of what's going on in industry in regards to Silicon Data Management and PLM. I'll put some great links for you in the show notes. Until then, we'll see you next week. Always remember to stay connected to the ecosystem. Bye -bye.