>> all right, we've worn you out. all right, so, if you have any other questions, especially late at night, if you would just please call greg, he will be our team scribe. but, thank you guys for listening, we appreciate, we'll be around if you have any other questions. i will tell you we are leaning on partners and friends and developing relationships here, obviously, as we look to work in that market in a new and different way. there are several companies that we're teeming with facilitating them entering a new market while we as a company work to enter a new market. so very much appreciate the collaboration. i thank you and enjoy the rest of the conference. [applause]. [inaudible conversations]

>> i wonder if i could ask you to take your seats please and we'll get started in our next panel. before we do, and while you're making your way to your seats, i have a special shoutout here from our team. one of our key managers from savannah river, kelly was promoted. so, kelly, congratulations. well done. [applause] >> all right, and i'm honored once again to be able to chair a panel of the three directors who probably have if not the most difficult job, one of the most difficult jobs in doe and nsa. and we'll talk about some of the challenges they face and this is-- we've done this a few times before, but this is the last time that james is going to be here with us as he's announced his retirement. we've tried for a year to talk him out of it, unsuccessfully

so. thank you all you've done for the country. let's have a round of applause for james. [applause] >> with that we're going to make this a real panel and dispense with the podium. so what we'd like to do today is talk about the same things that theresa talked about earlier today, the key elements of success. for nsa in general and the national security enterprise and of course, the labs. and these three lab directors and their institutions have made great progress in the last year in a number of areas, and what we're going to do today is talk about that and then talk about what comes next, and just let me-- by teeing that up and you've heard that in the briefings this morning and met all the key milestone from the life extension programs and the r

and d programs, streamlining the way we do business and all of them have had both of this bottoms up and top down approach where they challenge their team to get them innovative ideas and then they heard both from the new employees that we have and the existing employees and the ideas and made them actionable in always using a risk-based approach and make sure we're applying the right requirements to each mission and project. third streamlined decision making and they've done a tremendous job in that and this is the basic of moving the decisions down to the lowest possible level to let things get done faster. at the same time, maintaining the proper level of oversight and governance which we all know is critical using digital tools and other things, and having good metrics and good contractor assurance programs and of course, for the above using digital tools to analyze this mountain of data that we generate to detect trends and

inform the work that we do in real-time so we avoid the problems not just find them and fix them, but avoid them before they can impact our mission and projects. what we can do is talk about the same four teed up. first we'll talk about innovation and then we're going to talk about infrastructure. thirdly, we're going to talk about operations and then finally, what makes it all work is developing the work force of the future, our people strategy. so let me start with innovation, i've mentioned before implementing digital tools and digital transformation as areas where you've made great progress. can you describe what you've achieved? and i know, in the last year you've used it to improve our design methods, improve our operations and our production capability and of course, in maintaining a strong science base and then, finally, if you could wrap into that, what do you need to take that next

step? what do you need from nnsa to be able to take the next step in innovation? do you want to start? >> i picked the wrong chair. thank you, don. [laughter] >> that was a big question and thank you for bringing us back together again. we've spent a lot of this year talking about the digital transformation that we're trying to drive forward and i think it's going to impact our work in a number of ways, but i'll start with the simplest part, simplest, conceptionally simplest park which is the backbone to connect the scenes in a simple manner and we have more in common in the infrastructure that we can share seamlessly in between sites to streamline how we go from engineering and production to make those cycles efficient and modern, and provide what we're hoping will be a digital threat for the work that we do, so that we'll have really good data and great tools to enable

us to do modernization in new ways. within the laboratories, we're also using our new tools and capabilities. you heard earlier that we just cited several new computers. tom will talk about what we're doing at los alamos, and this is the conspires exo scale computer for national security. we're number one. which is pretty exciting. but the computing is now at a scale to do things that seemed inconceivable three years ago. and 3-d modeling is a design and linking together our engineering and physics design capability and 40,000gpu's in it, an amazing tool for things like ai so we're really starting to bring that ethos into our science and design communities to speed the pace of learning and really be able to think very differently about

how we do all elements of our work on the stock pile. so i think i'll stop there and pass over to colleagues. >> i'll just pick up right where kim left off, actually. you know, if you look at what went into making el capitan possible and by the way, all three labs are going to be able to use that resource so we're kind of basking in reflective glory. i think there are a dozen key technologies that went into enabling that tremendous computational resource that were developed as part of the exo scale computing project that are actually baked into all the big investments that are being made by the hyper scalers to train their models. so, it doesn't get a lot of attention, but that exo scale program actually was the building blocks of what is, you know, transformational

technological revolution that's underway right now, and so, you know, when you read about open ai announcing star gate and x-ai putting 100,000 gpu's working to gpu's in memphis. under the hood is technology at that came out of our decades of pursuit of ever faster computers. and one of the nice byproducts of that are machines like el capitan or the super computer that we deployed at los alamos in the spring are not only wonderful for doing the multi-about physic simulation critical to our ability to assure the effectiveness and safety and reliability of our nuclear stock pile. they're also going to be great for artificial intelligence because we have this kind of hybrid architecture and i think one of the things that we're really excited about is the possibility of adding that to

our tool kit. not to replace the modeling and simulations that we have, but to accelerate it. in fact, it will be an important training data base for ai and also, to take advantage of the tremendous resource we have in terms of experimental data and that's something where the three labs are really working very closely together, you know, we're hopeful that we can make an announcement of that before too long. but it is something that's going to enable us to go faster. and that's important because we are going to be asked to go faster. we are being asked to go faster, you know, the strategic posture commission captured that with the necessary, but not sufficient and we are very, very busy right now with things like infrastructure, topic to come up. and the only way we're going to have the capacity to go faster is to use some of the tools that actually we've been working so hard to enable over the last few decades.

>> well, kim and tom covered a lot of the pieces of the digital transformation, but there are some other areas here where i'll give you the positive pieces and then the challenges. so, in the w93 program it will be born digital. all of the integrated product teams have bought into basically using digital thread as doing their design through manufacturing. and in some cases the ipt's to digital threat. but it's not without challenges because there are eight labs, plants and five steps to come together and agree on a common set of tools and i would just tell you culturally as a challenge at sandia itself and coming with the ones that we're going to provide, design tools, only a certain version of it for our engineers. and then things as simply as staying up-to-date, which is something as simple as, you

know, our word processing tools, that they are all at the same version level. there are a lot of challenges to this and i believe tomorrow laura mcgill and jamie wolf will be on a panel talking about this and you can ask them more detailed questions, but the bottom lines, we're making progress in these areas, but there are a lot of challenges. >> can i ask you then to address under renault evaluation, a term that jim mcconnell uses and may have been coined by the nasa administrator, technology inversion. you've got a strong, tremendous technology base on all the labs and you develop these technologies and i know one thing that i've heard you all speak about is how long it takes us for a lot of good reasons and some not so good reasons to deploy them. what innovations, what progress have you made in the last year in technology insertion? you say when you can deploy those technologies sooner, it can work them into the mission to build faster and better and more efficiently, so, kim, do

you want to start again with technology insertion? >> sure, i think this is an area where we've actually made a lot of progress and as james mentioned with the digital transformation, changing the kind of technologies we use and the way we use technology is equal parts technical and cultural. it's not as simple as saying i have a great new idea or tool or technology, let's just roll with it. there's a high degree of confidence we have to build in a new technology or a new manufacturing approach and there are many legacy prophesies and procedures built around the way that we do things today. the cultural piece cannot be ignored and i think we have built a strong partnership that's allowing us to gain speed in starting new technologies. for us, we put a big emphasis on new manufacturing technologies and what's interesting to me, you usually think about designing technology and then building things, manufacturing the tool to enable you to do things.

the capabilities we have now with advanced manufacturing tools like additive manufacturing, mean it changes the way you think about design because you can make things that are very different from what you can concede about old tools. so we're sort of seeing the whole process many much more iterative as we learn the capability and the power of these new pools and things like machine part, inspection and control. and design optimization tools that allows us to iterate through many, many rarations -- iterations of a part. for the 87-1 mod, we're inserting new technologists because we have the capability now to change the way we've thought about the designs to make them more manufacturable, sustainable. easier to maintain through the life of that system. >> yeah, i think we've gone

through-- and as a physicist, we've gone through a phase transition. you know, there was a long period of time where really built into the way we were thinking about things with change as little as possible. we have designs that have their orbegins origins of testing and anything that we do for a change or how we manufacture for uncertainty that might cause us to question that basis and i think we're over that now. we're over it for a couple of reasons, first off because of the model and simulation tools, we have a much better ability to understand consequences of the change in material or the change in the manufacturing approach and actually convince ourselves that it's okay. i think the other thing is that-- and james owen mentioned this in his session this morning,

it's not just a question of can we design things that are easy to manufacture, it's what does the new manufacturing technologies enable in terms of design possibilities that simply could not be accomplished any other way? and the fact that actually with a lot of the, you know, added manufacturing and digital design complexity is much less frightening than it used to be. you can actually make things that are very, very complex in a relatively straight forward way and that actually gives the designers some flexibility in terms of how they approach solving a problem that they did not have in the past and we're no longer bound by that and i think we have enough confidence and ability to understand the changes that we can take advantage of it and also take advantage of the fact that it may require a smaller footprint, you know, it may produce less waste material in the form of spoils and the

materials we deal with, rather not deal with larger quantities of unused materials and all of these things are benefits that can help us, you know, stay within kind of a bounded cost envelope. >> let's see, i'm pretty excited right now because last week we had a successful 80-4 flight with the air force and there's new technology in there and of course, we can't go into specifics, but between lawrence livermore's technology that they've inserted and sandia, this is a really amazing advancement in how we think about these systems. there's a lot of stuff coming through, kind of laboratory scale that's coming through that's really going to change the way that we think about our safety themes for nuclear weapons systems. so i'm really excited about that coming through. >> so the next topic we're going to move onto number two and that's infrastructure and kim, relax, i'm going to start

with james this time. we'll start on the other end. so in infrastructure we know that some of our infrastructure was built 75 or 80 years ago, and need to be upgraded. and the enterprise blueprint that we all talked about and we all contributed to, it's a great description for the must-have and as the past administrator said, it's not a wish list, it's essential to the mission to support for the next 25 years, so with that, what are the key things you need at your lab to effectively carry out the enterprise blueprint, including help in implementing the creative approaches and you're all working on supply chains and standards broaden the supply base. do you want to start with that? >> i'm excited about the blueprint and having negotiated with the nsa and other labs, the sfis list of facilities

that need to be recapitalized and i want to go in a different direction for your question, john. because we've got to sustain a lot of facilities before the new ones come around, so that's going to be a really significant issue from the funding standpoint. we have to continue to have facilities that make the electronics for the weapons. we have to have the facilities that qualify parts for the weapons, and it's a-- if you look at the enterprise blueprints, some of the facilities are five to 10 years from now, before we get new facilities. and we put these facilities in a run to failure mode nearly a decade can go-- ago. that was a good decision at the time, and recap the facilities today and it's going to take longer so we've got to sustain these facilities to monetize,

and others that we need to pay attention to, lab space and office space, a lot of that needs to be recaplized through the complex. there's a lot to be down here, overall excited about the blueprint because it lays out kind of a timeline for the facilities that we absolutely have to have, but some of the facilities we absolutely have to have before they're recapitalized. >> and the important thing in my mind about the blueprint. it does look across the spectrum of types of infrastructure that it needed to support what we do, it's easy to spend all of our time focusing on the really big ticket nuclear facilities because, you know, they do take a tremendous amount of dollars to get built. they take a long time to get built and they have a very significant requirements in terms of safety and security

that drives all of that cost, so you see a lot of focus on discussions on things like, you know, ups, the plutonium infrastructure investments at los alamos and savannah river. and the blueprint falks talks about what we need to assess and they're scientific tools, their test capabilities that are absolutely important. actually, paradoxically, even though sometimes we take the shorthand to refer to these as our science facilities. if you want to know the shortest path between infrastructure and impact on the on alert deterrent, it's actually through those facilities because it's in resolving questions that may arise, and in the surveillance programs or whatever and to be honest the pits that we're building at los alamos as important as they are, they're

not going to affect the and alert deterrent until sentinel is in the field. so there are a lot of pretty urgent things in that scientific piece of the infrastructure that are ever bit as important as the big facilities. as james pointed out. one of the concerns is bridging, so, at los alamos, for example, we got approval, critical decision zero. the mission need for much needed modernization of the front end of our lance accelerator, which is one of the tools that we use to resolve issues, you know, important for qualifying new high explosives, for example. but we've got to keep the facility running, too. and in fact, this year, we're curtailing operations because we don't have enough spare parts. so while the lamp upgrade is tremendously important in ensuring the future of that facility, the more mundane things that are, you know, we

call it maintenance and operations and it's not as flashy as el capitan, but it's extremely important for the ongoing health of our deterrent. >> yeah, and now, i agree wholeheartedly with james and tom. one of the best parts of the enterprise blueprint effort was the time we spent as a community coming to consensus what's on that report, and so, what you read there is a shared commitment to the success of all the sites and i think that's an important place for us to be because the need is so large. at my site, you know, we have sort of three different categories of needs for infrastructure. first and foremost is this recapitalization of some of our scientific infrastructure. so we have the national emissions facility, our large laser facility which is a very high-tech place, but it's been operating full tilt for 15 years now and need for

sustainment of the facilities and recently received critical decision zero need to do an energy upgrade to the laser where we'll be able to increase the laser energy output from 2.2 mega jewels to 2.6, and maybe higher allowing us to push that facility and our ignition experiments in the more high yield regimes which is are critical important for our support to the stockpile. we use that facility routinely to do material testing and exposure and to really study in depth the science of nuclear weapons. so it's really an important day-to-day contributor for our spot to the stock piles. >> and in the second category, i would put enabling infrastructure so we have a lot of needs on our site as an example, one of our most recent line items was a power and computing upgrades to our computing facility so we could cite el capitan. that infrastructure is not

glamorous, but critically important to operating the facilities and our sites in efficient and sustainable way. and then for us, for those of you who haven't visited livermore, it's a small site. ... site, we are one square mile and we have 9000 employees. that work force has grown by more than 50% in the last 10 years. we need office space and we can't g ppr way out of this problem. the average dpp scalability is too small for a site where the footprint is we've been working close partnership to get a critical decision zero foreign investment in office space infrastructure so we can build up a little bit and much more efficiently use our size and bring our weapons program teams closer together so we can foster, collaboration, for important work we have to do going forward. it's been an interesting journey. we have beenn working on our

infrastructure plans with our partners at thehe production sites. also trying to build out production development capabilities that allow us to study the site off production ad help bring new capabilities for production of polymers, explosives, and other key components to aid this process of modernizing in the production facilities. >> thanks. we talked earlier today and it's a great thing to celebrate 75 and 80 years at our facilities. the real celebration we all want to happen you hear this from plants and sites to mark is where you and roger and eric and rich and kelly when you guys commission those new facilities you are buildingti now, that's n be something to celebrate. dj, but that down for your list of champagne toast for the future if you would please. let's cut operations then. as we expand operations to support the upgrades for national security capabilities,

what are your top issues you need to address? obviously we've been doing the research. we talked about operation and infrastructure but as when the board operations, what are the challenges you face? power using digital tools, to ensure the strong, safety and security performance and the performance of this new facilities? p overall, what keeps you up at night when you look at operations, moving into the phase which is exciting and essential, but what are the things that keep you up at night to say these are really the things we've got to address first and foremostt to make a successful?sf >> certainly our biggest challenge now that, in fact, at the event where we were celebrating receiving the diamond stamp, first production unit, it was pointed out that this is not a finish line. it's the starting line.

we've got a lot of work ahead of us over the next couple years to build out the capability to get up to the 30 pits per year that teresa mentioned in her remarks. we have to do that in an operating facility. that is the challenge because we want to continue producing pits, maintaining the competency to produce the pits, train people to be proficient in doing that work at the same time we rethink out obsolete hardware and bring in new glove boxes, and that's in a nuclear facility that has been operating for decades, and doesn't is as early have the best as built drawings and occasionally when you open a valve to find their something sitting inside there has been around for a while. that's the quickest way to knock

us off our pace in terms of the infrastructure build, is an operational offset that would bring work to a halt. if it's not safe, when a a dog any work until we remedy that situation. i think for us the biggest operational challenge is actually the interleaving of the operations with the infrastructure work. we don't have the lecture of a greenfield location. there are other important missions in pits and their ongoing as well. there are areas in porn surveillance were quick to do. there's the t heat sources as wl as r&d. it's a kind of unique challenge and it's one that keeps me and actually a large number of other people up at night. >> james, do you want to go next? >> the things that keep me up at

night are the maintenance of these facilities. i'll give you a couple examples. at the angle core reactor which is a critical facility for qualifyingil hard signal nuclear weapons, we lost a safety rod and it started to leak. fortunately we sought. it wasn't an immediate issue but we had replaced that broad. there was only one spare. wepl are down to no safety rodso we need to build up some new ones but they kept the reactor down for a year. that's the kind of things that happen. it's typically at leastpp a year speeding we believe this with my beat up what all of our programs online at c-span.org. live now to the democratic national committee winter meeting being held outside of washington, d.c. publisher from outgoing dmz chair jaime harrison, ahead of party leadership elections. live coverage on c-span2. >> she's not running for reelection this