Now here's a new concept. Digital network advertising for businesses. Display your ad inside their building. If a picture's worth a thousand words, your company is going to thrive with digital network advertising. Choose your marketing sites or jump on the DNA system and advertise with all participants. Your business ad or logo is rotated multiple times an hour inside local businesses, where people will discover your company.
Digital network advertising DNA a novel way to be seen and remembered. Digital network advertising with networks in Redlands and KAIPA call in the nine on nine are two two two nine two nine three for introductory pricing. That's nine O nine two two two nine two nine three for digital network advertising. One last time Digital network advertising nine O nine two two two nine two nine three. Tahebow Tea Club's original pure power Darco Supertea helps build red corpusals in the blood,
which carry oxygen to organs and cells. Our organs and cells need oxygen to regenerate themselves. The immune system needs oxygen to develop, and cancer dyes in oxygen. So the t is great for healthy people because it helps build the immune system, and it can truly be miraculous for someone fighting a potentially life threatening disease due to an infection, diabetes, or cancer. The tea is also organic and naturally caffeine free. A one pound package of tea is forty
nine ninety five, which includes shipping. Two order. Please visit tihebow Tea Club dot com. Tihebow is spelled tea like tom ahee b like boyo. They continue with the word t and then the word club. The complete website is tohebowt Club dot com or call us at eight one eight six one zero eight zero eight eight Monday through Saturday, nine am to five pm California time. That's eight one eight six one zero eight zero eight eight to heebow Tclub
dot com. With sixty years of fascinating facts. This is the man from yesterday, back in time, this time nineteen fifty seven. ABCTV says it's excited about another potential young star and when to debuts Maverick in September. The series stars newcomer James Garner as a crafty, traveling card poker player Maverick. And from this time in nineteen ninety four, CBS Entertainment's going to debut a new show, Touched by an Angel with Roma Downey and Delaese. Della Reese
is known primarily as a singer. Roma Downey is best known for her portrayal as Jackie Kennedy Onassis on an NBCTV mini series, There's nobody here, Nick, Maybe you've had enough? Huh, he can't see me, Nicholas, because I'm an angel. My name is Monica, an angel angel name And from this time in nineteen sixty three, MERV Griffin says he has a new game show idea. Mc griffin says he's been fooling around with the format for
about six months and he just about has it down. The name of that game show will be called Jeopardy. Hello, here's the star of Jeardy. Arn't wunning. Good morning players, Don Pardo Dag you my friends, Good morning. I hope it's a good one. Boyant Dagger with Moore at Man from yesterday dot Com. It's time to make the Tri City Center in Redlands a regular part of your weekly shopping experience. Try's City is home to a
wide assortment of quality businesses, including the all new Ocean Aquatics. Check out their variety of exotic tropical fish along with fish food accessories and tanks of all shapes and sizes. The Tri City Center is located just off of Alabama and the Tennessee exits in Redlands. Visit the Tri City Center today and find out why it's called the Mall with a Heart. Do you like to safely leverage bank money to earn double digit returns income tax free, with guarantees and no
downside market risk? How can you do this? This is Fern's host of the Your Personal Bank Show. One. You fund a high cash value policy one time to earn dividends and interest to establish a bank line of credit using the cash in your policy as collateral. When you earn more in dividends from your policy than the interests the bank charges, you keep the difference, and the differences average two to five percent annually in your favor for the past forty
plus years. Three the bank funds contributions years two to twenty plus. Each year the bank adds funds, your rate of return increases. Your average raid return can grow to strong double digits annually within a few years. Contact me at your Personal bank dot com your personal bank dot com or eight six six two six eight four four two two eight six six two six eight four four two two for more info or tune in to the Your Personal Bank Show.
Your Personal Bank Show airs Tuesday's at four pm right here on Case a A ten fifty am and one oh six point five AFM, this station that leaves no listeners behind. Join me Stephanie Miller every weekday for the antidote to cantankers, conservatives and if they don't like it, too bad. It's on Caseyan weekday mornings from seven to eight ten fifty am and one oh six point five at m. E digits lock them in for more information, recreation and guaranteed
fun k c AA. The information economy as a rod, the world is teaming with innovation as new business models. We invent every industry industry. Inside Analysis is your source of information and insight about how to make the most of this exciting new Eric. Learn more at inside analysis dot Comside analysis dot com. And now here's your host, Eric Kavanaugh. People. All right, ladies, gentlemen, welcome to the future in ead your host Eric Kavanaugh here.
What a wonderful show we have lighted up for you today and the only coast to coast radio show all about the information economy. It's called Inside Analysis. Got my good buddy Ri Christian from sixth Sense Advisors. We're working on some cool stuff these days, and we have Chris Gladwin, the CEO of a company called Oceans that is fundamentally changing the data warehousing landscapes. So we're going to dive into what that all means for you, how it's going to
help you get your business done in the near future. And really I wanted to talk about the evolution of data warehousing. So Chris and I are old enough to remember all the different eras we've gone through. It used to be that a data warehouse took you two years to stand up, or sometimes a year if you were very good at it costs you millions of dollars. Don't even bother trying unless you have that kind of expense. And then things started
to change. We went through this whole had dupe world that we've talked about on the show many times, where people thought, oh, no sequel, is that not only sequel? Or is that no sequel? They thought that map reduce, which is how Yahoo index the web, could be used to do a lot of fun stuff with data, and they did. There were interest some interesting use cases that came out of that. But then cloud storage dropped precipitously in price, and that kind of killed the value proposition of hadoop,
which was already a bit of a cluji environment. And then what happened, We had this massive resurgence of sequel. So, for those who don't know, sequel stands for the structured query language. It is the lingua franca, if you will, of database query languages. It's how you query a database to understand what's in it. And this whole search came out with a company called Snowflake, which at the time and I think had the biggest IPO in this industry. Ever, now that's come back down to earth a bit.
And then you've got this company Data Bricks that comes along. Then you've got Amazon Redshift. There are all these different ways that you can slice and dice data and do your data warehousing, so what makes sense for your company? And then this new company, Ocean comes along, not exactly new. They're on version nineteen a while ago, I think, then version twenty one now, which shows you the maturity of the platform. They came up with
something called a hyper scale data warehousing. So what does that mean. Well, the cloud gives you this hyper scale capability. Right, So people started to understand that Google and Amazon and Microsoft, they're the hyper scalers for being able to deliver functionality at tremendous scale that can scale up increasingly, you can scale down. That was a bit more of a challenge than scaling up. But what does that mean for data warehousing? Well, we have demand to
explain all of that. Dialing in from my old old heart of Lamont, Illinois, Chris Glabin of Oceans, tell us about your vision and what is hyper scale data warehousing? Hey, Eric, great to be here, Chris, great to talk with you. Yeah. So, hyperscale is a term that I think has been used at different times because it changes in its meaning. What we're seeing is what hyper scale means in data analysis is it's the
point at which the scale of the data analysis dictates a different approach. You can't just take the normal kind of middle of the road and say, well, let's just make it bigger. There's a point at which the scale of
your analysis benefits from an approach just for hyper scale. So that's one thing we see and where we're seeing that in the market is generally when the average amount that you're analyzing is about a petabyteed data, So every time you run a query, it's got to look at a petabyte, and you do that
a lot, and you're doing kind of complex things. And in terms of numbers, what we're seeing is we have gone through the process of identifying what are all the data sets in the world with business requirements that need that, and we're seeing it's about on a dollar basis, it's about four percent of the overall two hundred ten billion dollars data analysis market, so it's about an
eight billion dollar market growing really fast. It's growing not only because the data keeps growing, but the number of use cases that need that approach is growing, so it's growth on growth. So it's growing about thirty five percent a year to about thirty five billion dollars in just five years. So that's the mark. And then you get into like what's in there? You know, what is the kind of data that needs this, And certainly we're seeing the
biggest source of that is telecoms themselves. So if you're a telecom your network is giant, and there's all kinds of reasons why you want to analyze what's going on. It could be compliance requirements you have, it could be capacity planning, troubleshooting, performance optimization. But to really do that, you've got to look at all the metadata that flows in your network to see what's happening.
And the amount of metadata that it's telecom makes is crazy. It's not trillion scale, it's hundreds of trillions, and quadrillions is the next number. You know, that's actually what they want to analyze. So that's just a little flavor of kind of what hyper scale means. Yeah, you big up an excellent point, because there are these inflection points in the industry, right and so historically you'll see things like that's where appliances would come into play.
Right when the software can no longer handle it by itself, you do an appliance, and then you're leveraging the power of the hardware and the software. And then of course that kind of changes again, and so you innovate on the software. And as my business partner, who's mostly retired now, Robin Blore would once pointed out, all the innovation starts at the hardware level.
At some point, some new piece of hardware changes things. And we were talking before the show about solid state memory, and if you look at there's a lot of interesting threads we could pull here. But you look at Osso Platner and this is a bit off topic, but he saw, like twenty odd years ago, twenty five years ago, he said, you know what, memory like solid state is going to come down in price over time, and at a certain point it's going to cross with data stored on disk.
And so they pushed SAP into this whole in memory world. Well it's before kubernetties came out, and so now they're you know, kind of sitting on this monolith in a new distributed world. It's a bit of a different challenge. But he saw that coming in. You've kind of seen that coming too, right, because the capacity of these flash and these solid state drives to
deliver data much much faster with less trouble is a big deal. So but figuring out how to build the infrastructure to leverage all that, well, that's what you guys been working on for twenty odd versions, right, Yeah, And we saw that coming almost exactly ten years ago. The revolution that's happening in solid state is not is not wasn't. It wasn't something you could hide because it literally was close to fifty billion dollar investment made by semiconductor manufacturers like
Mintel and Samsung and Tashiba, like that doesn't hide in a corner. You know, that's like ten billion dollars fabs and thousands of engineers and you know, lots of academic papers and it took a long time. And so one of the things that it's different though about this what I would argue, it's the first new building block and computing ever. You know, yeah, your hard drives and they got bigger and DRAM got bigger, but solid state it's
not a faster spinning disc. It's not cheaper DRAM. It's different, and you know, you could see it coming from from you know, a decade away, and unlike most of the time or I think pretty much all we have, I'm curious what you all think, but I would assert that this is the first new semiconductor change that didn't start like in supercomputers and work its way down with the opposite. Apple as a phone manufacturer, device manufacturer,
and Samsung also saw this coming and they got in the production. You know, they were investors in the fabs and the fab companies, and part of that was like I get first crack at it. So what we saw as you know a company completely dedicated to building data analysis platform on solid state and we have friends. I mean, we've been doing this for a million years.
And there was a point where it was in your phone. It was in like billion unit production scale in consumer product the phone, the laptop, and we as the leading technology creator for hyper scale analysis platforms, we could barely get our hands on samples. It just wasn't in production like consumers were first. And then what happened and then you had the supply chain issues with COVID where you know, it just made it really hard to get something unless
you had a guarantee contract. And then about twelve months ago it changed. Now you want a million, you know, no problem. You know those fabs are running, they've got the production issues. And the profound change that you were referencing is what's happening now is it's gone from really hard to get your hands on so you can get as much as you want. And there's two metrics that matter. One is the cost per storage and the other is
cost per performance. And what is happening right now is if you look at the life cycle costs and include only the cost of the drive, but the cost of power because they're more efficient on power than spending disks the space which costs money to put in a data center, and they're more efficient on space than disc. And reliability. When something fails less often, it costs less
in solid state fails less often than spinning disks. If you look at that life cycle cost solid state, you know, particularly MBM E solid state, which is the parallel interface that everyone uses, it's a great standard. Is crossing below spinning disk right now on a cost per terabyte basis with currently two thousand times the performance, and the next generation will have four thousand times the performance, then eight thousand times the performance, and spinning disc will never get
faster. So that's happening right now. And what that will cause, and it will first happen at hyper scale, is the collapse of computing storage tiers into one. And the only reason they were separate because there was a big
price difference. And then what we can talk about maybe later is around twenty thirty the cost per performance, which you know, we're using the cost per million random four k reads per second per dollar solid state will drop under d ram under the cost for performance, you know, and so and salid state is so early and it's Moore's law curve that it is in Moore's Law. Every eighteen months it gets twice as good. DRAM is not, spinning,
disc is not. They're kind of stuck. And so this is where we are today and where we're headed as solid state is going to completely change, you know, how computing is built and how data is analyzed. Yeah, that is absolutely fascinating. I'm going to bring in Chris Christian to comment on
that. I mean, these are the kinds of things that engineers need to be focused on, and frankly, CDOs and CTOs and CIOs need to have in their perspective because this is going to have tremendous impact on what you're able to do if you're not embracing this new way of doing things. But Christian, what do you think? Oh? Absolutely agree? In fact, oh highlight where Chris was taking the thought process too. If you want to think of in the next seven years spectrum twenty thirty, right, what's going to
change? The question is what's going to change? That's what bothers a lot of CIOs and cd aos, not just the digital officers, but the data and what's going to change. We are becoming drawn driven Everything is getting automated. Everything is becoming its own landloans and its own points where data is going to intercept and criss cross, and what boundary is safe versus what is okay? Verse public versus what's private? When you're looking at those boundaries exploding.
In the next seven years, we're going to look at volumes of data, which is like, yeah, I built a five terabyte data whereas people will laugh, five terabyte terabyte would be like something that happens on a daily basis. I built a five beta byte data ware Else, now you're going to be talking, yeah, that's good, But then we're going to be talking about fifty beta bye data warehouses. But we'll still say data warehouse, right,
I mean some tween. While the spectrum of storage is fantastic, the problem solving still has to be put on top of it so that you can tether and then get the solution stack built on top of that foundation. And to me, that's where Eric, I think we should take our next set of discussion points with Chriss. Why datawareousing? Yeah, well that's a good
question, So go throw it over to you, Chris. Now you have these data lakes people are talking about these large language models doing very interesting things, but it's not data warehousing for sure. And you're really focused intently on the data warehousing use cases at hyper scale to enable organizations to really slice and dice, right, because the thing is the factor you can cut through the
data, the factor you can do the analysis and get the answers. The more questions you can ask, the more chance you have of finding the signal you wanted to change your business, right, And I think a lot of the older solutions now are frankly encumbered not just by bureaucracy, but by technical debt and sort of traditional hurdles that they're just not getting over. They're kind of running into walls. That changes behavior inside the organization. You ask less
questions, you're less risky. If you will, you take a fewer chances, and then you don't do as well. I mean that's my take on it. But what do you see, Chris. What we're seeing in customers is the compromises they've had to make at hyper scale often involve things like, you know, letting most of the data fall off the table. I mean, they just they can't effectively analyze it, so they say they don't even collect it. You know, automobile manufacturers, if you have one hundred million
cars. You know, I talked about tell for a telco prior to oceent. In this next generation of data warehouse capability that we're talking about, they really couldn't cost effectively collect and analyze all network metata on the network. It's quadrillion rows on a table. It just doesn't make sense. They're not going to spend half a billion dollars on a supercomputer, and so therefore then they end up making compromises. So if you want to do like performance analysis,
you don't care about averages. All you care about is the worst. You know, what's the worst performance on my network and what were the conditions. But that means you actually have to have all of it to see the worst and identify the things that correlate with the worst. So what we've seen where you know, companies would then either just not have all the data and they make the best of it. In some cases like an ad tech, some
of our customers they would down. You know, there's there's you know, tens of millions of digital ads that go up for auction every second, and there's whole industries of people that either supply those inventory of ad placements or buy those and analyze those. And because you know that's such a volume, if they want to back test an algorithm for some kind of campaign they're going to run on the last three months, well that's just too much data, So
then they would downsample. It's something like point one percent of that ad exchange. But the net of the problem with that is then if you had a really large customer that wanted to test a campaign, the results become inaccurate. So what we're seeing is because attociate, we're using this solid state technology, which is like game changing. It's you know, and going back to your time in earlier, the best you can do as a software designer is to
go as fast as the hardware. And so what we're able to do is to address these problems where you can do full resolution data results can be accurate even at hyper scale. Yeah, that's fascinating stuff. I mean really, and I'm sure there are some mindset issues that companies have to kind of get over. But the person in the room who understands the knowledge is the person who is run into these barriers and to your point, has had to say,
well, let's just use point one percent of the data. I mean, down sampling always gets you that lossy characteristic right where it's like, well, am I now going to lose visibility into the signal I was trying to find in the first place. Well, that's not a very effective solution.
And again to be able to do it quickly. And what you mentioned the stat like two thousand times fast or some of these these new solid state drives are compared to their counterparts, Well, you know that's more than an order of magnitude. That's more. That's that's two orders of magnitude, right, It's like just off the charts kind of and it fundamentally changes the conversations that
you have and how your team operates. And so that's what excites me is you're opening this new portal for companies that go down the hyper scale highway with data warehousing to where you don't have to limit yourself to small views of the world. And that is going to fundamentally change and it saves tremendous amounts of money. If they can gather all the metadata a telco and do your real time analysis, you could wind up saving millions, hundreds of millions of dollars
just in not wasting money, not wasting time, not wasting energy. Folks will be right back to tech doll you're listening to inside analysis. What if you could own a piece of the future. What if you could build your next castle not on sand, but on the bedrock of a modern blockchain ecosystem. The first Internet gold rush made millionaires, The second wave is minting billionaires, but the third wave is just gathering now and anyone can get in on
the action. Hop online to crowdpointtech dot com to learn how you can secure a foothold in the blockchain revolution. Whatever your passion, wherever you want to go in life, there's an opportunity awaiting you right now. Go to crowdpointtech dot com to learn how the blockchain will fuel the next generation of innovation in this globally connected world. That's crowd pointech dot com. Your trusted agent in an untrusted world. What's the longest running radio show in the world focused on
data? DM Radio? Want to be a guest sometime? Send an email to info at DM radio dot biz. That's Info at DM radio dot biz. Can your IRA stand up to the next financial crisis that our top economists are saying is at our doorsteps. By allocating a percentage of your IRA into physical gold and silver. With a tax free rollover, you can diversify in safeguard your holdings from turbulent markets and economic downturns by putting your IRA back on
the gold standard. Find out how to safeguard your SS with a tax free rollover with a Genesis Gold IRA, the only IRA that can hold physical precious metals. Call now for your free gold and silver report. Protect your IRA today with one simple phone call and learn how to qualify for up to ten thousand dollars in free silver called Genesis Gold Group Empowering Faith driven Stewardship. Eight hundred six four four eight six one one eight hundred six four four eight six
one one eight hundred six four four eight six one one. That's eight hundred six four four eighty six eleven. Do you own a timeshare? We'll face the facts. You made a mistake, you made a bad purchase. A timeshare is not an investment. It's a money pit that continues forever. If you use your time share, that's great. But if you don't and you want illegally get out of your contract, call my friends right now at the
timeshare Exit Hotline. They're an experienced team of lawyers who help good people like you get out of a timeshare contract that they just don't want. Don't throw away your money on maintenance fees. Use it for things you really want. We can help you end your time share contract and stop the money drain immediately. If you are ready to move on with your time share, call our
team right now. Le your time share now with a free call. Eight hundred two eight nine O four one three eight hundred two eight nine O four one three eight hundred two eight nine O four one three. That's eight hundred two eight nine zero four thirteen. If you served in the Marine Corps by now, you know about the contaminated water problem at Camp Lejune. If you were stationed or worked at Camp Lejune from nineteen fifty three to nineteen eighty seven,
you probably have a lot of questions. We have some answers. You could be entitled to compensation. Billions of dollars are being allocated to pay for damages to anyone stationed at Camp Lejeune during that time. Unfortunately, it appears that officials may have known the contaminated water problem existed and did little to protect their men. The Semper five code was not honored. If you or someone in your family has developed a serious illness, including various two five, four,
thirty two eighteen. Welcome back to Inside Analysis. Here's your host, Eric Tavanaugh show. All right, folks, welcome back to Inside Analysis. You're host here, Eric Kavanaugh talking with Chris Christian and Chris Gladwin of Ocean's. Chris's my consultant friend from many, many years ago, also in the Chicagoland area. I grew up in the Chicagoland area. And you know before the show, we were talking about three fifty five and how I three fifty
five opened up. And I used to be the editor for the Lamont Metropolitan newspaper and I actually interviewed a guy from the Illinois Department of Transportation about the project at the time, and he told me something very interesting. He said that the laws do not allow you to go and build a highway according to future traffic speculation, like how much traffic you think is going to be in
the future. They say, you can only use the data of traffic that is currently in this region, and that's how you justify where you build these highways. Well, once they built three fifty five. I promise you there are all kinds of people that change their driving habits because it was so much easier to go up to the North suburbs and it was like night and day
basically. And you can kind of make this comparison, this loose comparison of traditional versus hyper scale, and traditional is going the back roads and hitting stop lights wherever you go. It takes you an hour and ten minutes to get from Bowlingberg to Schaumberg, and then the hyper skill shows up. It's like, oh nope, you just get in the highway. There's no turn offs. You can just keep rocket and rolling and get there and no time.
It changes behavior and new things happen, and you start to change how you planned your day because I don't have to spend an hour in ten minutes. I can only spend twenty seven minutes. So these new technologies, Chris, when they come along, they open up whole new possibilities. So it's not just that you can do the old stuff ten times faster, or it's that yeah, you can do that, and you could do all this other interesting stuff, and that's where the business value tends to accrue. What do you
think, Yeah. Generally, what we're seeing is by designing a new software data analysis engine for hyper scale analysis using this amazing new building block, the NVMME solid state drive, as well as some other pretty choice building blocks like high corkunt CPUs and hundred GIGNIT connections, everywhere, you really can't get to ten times the price performance than what was previously available. In some cases you
get to a hundred times. And every time, you know, and this is not the only time there's been ten times the price performance for some kind of IT capability, It's happened many, many times. And every time it happens, two interesting things occur. One is industry or consumers never keep doing what they're doing before and just save the money right never. They always find if it's ten times the price performance, they will find more than ten times
the things to do with it, and the pie grows bigger. I challenge you to, you know, give me a counter example, because I'll name everything else. So that that always happens. And the other thing that's interesting, and part of why that happens is when you look back ten years, twenty years later at what are the uses now of that disruptive new technology typically at least eighty percent are new things that could only exist once the disruptive technology
change occurred. You know, twenty percent or less are what was there before becomes bigger, faster, cheaper, and more than eighty percent becomes only now that I have broadband? Can I have a giant system that distributes videos of pets known as YouTube and other silly things like that didn't that couldn't happen before broadband. You know, broadband happens boom, here comes YouTube, and there's
you know, many many examples of things like that. Like we in my last company, clever Safe, we dominated the market for photo sharing, for software for photo sharing systems. Photo sharing was kind of the first incarnation of cloud storage. It could you know, there was you know, this idea of like take as many pictures as you want, will storm forever and every once, so I'll sell your mug like that business cannot exist and that's a
very great business for a lot of companies. It couldn't exist until the price performance of you know, super reliable photo storage systems at hyperscale went down about one hundred times and you just you know, you just see example after example where when these big disruptions occur, all kinds of new innovation happens. Yeah,
now that's exactly right. And I love the comments you made of you know, when you do get this increase in performance, never did they just save the money, especially in analytics, Chris, because in analytics, again you're trying to understand your business. And so now when you have these new tools to dig deeper into your own data to understand, like you take the
Telco example, that's a big deal. If you can problem solve faster and figure out how to fix this issue, you're not going to lose a thousand or two thousand customers to your competition, for example. And just little things like that really add up quickly. But it's the case across the analytical spectrum, the more you can analyze, the better off you're going to be.
And you know the fact is that some of these old solutions, even though Territata, for example, I'm told, just spend two hundred and fifty million dollars to refactor their entire engine to run in the cloud and Amazon Web Services now they just rolled out for Microsoft recent it's a lot of money to spend
refactoring. But Chris has some interesting thoughts about Kubernetes as well, because with kubernetties, you're federating all this compute, which you know, I've always joked we sacrifice state at the altar of scale to be able to do these things. But still there are inefficiencies baked into that system as well. So if you do your job right on the ocean's side, there's really not a lot of benefits to get from going into a cuban, righties environment. All you're
trying to do is enable your analysts to crunch data. Chris will throw whatever do you first, and then we'll get Chris. Yeah, And the point you are bringing about the availability of valid data is very essential, right.
I mean, for many years we've all gone in the traditional route where we've said, yeah, this model work, said no, this one doesn't, or while a bridge table suddenly becomes a point where data just goes Christ all kinds of issues that have been faced for a number of years where analytics could not succeed because of the lack of data. The bottom line, right, that is something that I would love to hear Chris talk about because hyper scaling
would mean now your data fabrication. This is a term that ocean would love
to talk about, but a good definition. So I think if we can get give us a quick definition on what the data fabrication thought processes and then we talk more about why, then bringing the analytic suitcase or anything else is going to be so much more easier Or folks don't understand, but that's the bottom line, that's the grassholds that we want to get to, Chris or I think the other big change we're seeing in data analysis kind of from a
traditional even the kind of recent generation of data warehouse life hopefully can data bricks and redshift is a shift from kind of traditional batch oriented data loading to data never stops streaming at scale and in the database now has to exist in this environment where data is constantly growing, and that you know, one of the challenges is taking you know, terabits per second of input data which is never in an easy, easy to digest state, in getting that transformed from some
typical semi structured at messy state into relational schema that is index secondary index, compressed and cryptic and showing up in queries within seconds. Now that traditionally has not been possible either. And that's another thing that we were able to do it did Ocean because the big it's also interesting to think about where does data come from? And at hyper scale, zero percent of the data comes from
typing. People don't type at hyper scale. Just to give you a sense, a minimum system for Ocean would be, you know, a petobyte a trillion rows and a table a trillion if you print it out circles of the Earth seventy three times, and if you were to scroll it, you know, two pages a second. Assuming you could see data that fast, it's centuries to just scroll through that table like that is not something that humans can create, and that is not something that humans can even read, you know.
So a couple changes there, and that all the data being analyzed at hyper scale is machine generated. It's routers, it's cars, its satellites, it's instruments, and then the other big changes we've entered this time. I mean, it used to be like, you know, if you really had to look into something, you could go look at the locks. You know, you could you could actually go look at the data. Like you can't go look at the data anymore. Human beings cannot see data at this scale.
They are now completely dependent on these tools that look at the data for them, and an increasing amount of what people are using is machine learning artificial intelligence to characterize that for them. I mean, I can't look at everything, So just go look at these hundred trillion things and tell me what's forced standard deviations out of normal? And maybe I can look at that, but I can't look at the raw data ever. Again, So those are a
couple of giant changes we're seeing. That's funny. You have a couple great quotes that I can look at. They're on data. Ever. Again, that's a funny thing. And let's talk about observability too, because you were just kind of referencing that observe ability is this huge space now that has blown
up for lots of different reasons. But to your point, with observability, you've got all these signals of data that streaming constantly, and what you're trying to figure out is how to apply the appropriate filters to get signals when necessary. So you mentioned like four levels of deviation for example, can you talk about how you're able to leverage data warehousing technology at hyper scale in order to
enable meaningful observability across these systems. And there's also the issue of cloud environments, which have lots of different things going on. It's not just your ERP you have to worry about, or one database. It's lots of different things interacting in real time to push this stuff forward. So talk about how you're able to help people understand what's happening in that world and how that actually works.
An increasing amount of what's happening is what's happening right now. You know, it can't just be oh data it was, you know, fresh from yesterday. It's got to be fresh from three seconds ago. And so you know, so much of what you want to understand is involves this fresh data. So it's and it's not. It's sometimes just looking at that data itself, and sometimes it's looking at that data comparison to older data, but it's
less and less just looking at older data. So this real time analytics capability that we've incorporated at hyper Scale and our platform is essential, and all of our customers are using it now because all of them have, you know, this need and the big unlock we're seeing in there, just like we're seeing in the data. Analysis at hyper Scale is using machine learning more and more to tell you not only what is in the data that's pouring in all the
time. But what but how to deal with it? You know, it's takes you know, they like there's so much mood coming in, it's changing so much that we're using AI more and more to help us load and characterize the data as it streams in classification basically, right, So you're trying to classify in real time and ascertain what's normal and what's not normal. I mean, I think that's basically the rule of thumb. Right, this is normal
behavior. Now if something gets out of black, that's abnormal behavior. What's going on there? And this is how you have to figure out how to use these different filters to classify and to understand in real time. And you're you know, so let's think about Larry Ellison, for example, who I guess five years ago was talking about the healing the self healing database, right,
that was their big thing. It sounds like you're kind of going down a similar path where you're using machine learning an AI to ascertain, Wait, something is wrong over here, let's go fix that, Reset this, no, do this, do that? Whatever? Is that kind of what you're talking about, is this self healing kick? Yeah, And I would say that that self filling thought from five years ago kind of the universe there is that this database is a fixed sometimes fixed data set, and that's where you
have to heal. What I'm reporting is that the world has changed to where so much of the part of the database isn't just what's there, but what's arriving, so that the healing that you have to do isn't I mean, the static part. It's actually pretty easy. It's a lot harder to deal with, like what's going on now and has that changes? Because the world,
like it doesn't matter what the data set is. If it's you know, vehicle data, there's road construction, you know, there's traffic, there's a public safety and so if it's you know, ad tech data, something's happened in the market, So like how's that change, you know, the buying and selling of ad placement? If it's interesting, you know, I mean it just like you can't remove the dynamic nature when you think about anything, even you know, including just like what does it mean to heal?
What's the kind of harm that you cant fix? Well, it's it's more in the flow of data as opposed to the batch of data. That's very interesting, and that's true, right, I mean, data is always moving. It always has been moving. But you made a good point because the old mindset was Okay, here's the static date base that I have and something broke, let me fix this static things like, no, this thing is moving constantly. It's always moving. So you're trying to really understand the flows
of data. What is the nature of this flow of data in this particular environment, and point that over to like you say, some events, Well, what happened, Let's get to the bottom of this. That's really a magical characteristic of these modern systems. What do you think, Chris, go ahead, no other piece to what Chris was saying. I was. I think it was about twenty months ago. I was working with the San Diego
Department of Biological Sciences. They said they were trying to capture shellfish brain movement a bit. They had an issue. They said, we don't know how to make this effective. And I spent like almost a week trying to understand what were they trying to get? To Christ's point, right, I mean, if you don't know what you're trying to get, everything is going to be like, oh, it's outside this. Once we ascertain what we were
trying to get, in that constant flow. We could then say where the spikes were, where the abnormalities were, where there was an interception that needed to be done. All kinds of pieces could be reacted, but you need to have somebody who knows the system really well in order to understand what is it that you're missing so that you can take care of it at that very
minute. I mean, it's a latch on a sap. There is no Yeah, it can wait for something to come now, it's like, do it good, dare I'm now really understanding why Chris uses this term hyper scale and the on a past show we get people to find hyper scale and one guy said loss of control. And that's kind of what you're tie. It's not moss, but you just have to think of things differently. But don't touch the down books with your right baculistening fee. Do you own an annuity
either fixed rate, indexed or variable? Are you paying high fees and getting low returns? If so, Annuity General would like you to have this free book to learn the pitfalls and mistakes of buying an annuity. The Annuity Dues and Don'ts for Baby Boomers contains the little known truths about annuities like how to help reduce your fees and increase retirement income. And it's free. That's right free. As a bonus, we'll also throw in a free annuity rate report
just for calling. We researched over one thousand annuities and summarized rates and benefits from financially strong insurers. You get Annuity Dudes and don'ts for baby boomers and the Annuity of Rate Report, both absolutely free for calling Annuity General Today. Hurry supplies are limited. Call now eight hundred two four five one six ninety seven eight hundred two four five one six nine seven eight hundred two four five
one six ninety seven. That's eight hundred two four five sixteen ninety seven. Do you own a timeshare? We'll face the facts. You made a mistake, You made a bad purchase. A timeshare is not an investment. It's a money pit that continues forever. If you use your time share, that's great. But if you don't and you want illegally get out of your contract,
call my friends right now at the Timeshare Exit Hotline. They're an experienced team of lawyers who help good people like you get out of a time share contract. That they just don't want. Don't throw away your money on maintenance fees. Use it for things you really want. We can help you end your time share contract and stop the money drain immediately. If you are ready to move on with your time share, call our team right now. Castle
your time share now with a free call. Eight hundred two nine oh six seven O five eight hundred two nine oh six seven O five eight hundred two nine zero six seven O five. That's eight hundred two nine oh sixty seven. When a player's sudden cardiac event brought a national football game to a halt, it's shown a spotlight on the importance of CPR readiness. Now, with youth sports in full swing, the American Heart Association is rallying parents and coaches
to be ready in an emergency. To be ready, learn hands only CPR. It's a skill anyone can learn in minutes. Just visit Heart dot org slash hands only CPR. Hands only CPR is nationally supported by an Elevant's Health Foundation. Each year, three hundred and fifty thousand Americans did from a cardiac arrest. When seconds matter most, CPR can be the difference in whether a friend or family members of that's why the American Heart Association is challenging every household
to elect at least one person to learn CPR. If you have ninety seconds, you can be your family CPR hero. Just watch the American Heart Association's hands only CPR video at heart dot org and become a hero. Do you need to get your hands on some extra money right now? Maybe twenty five thousand or more If you're a homeowner, Now is a perfect time to get cash out while homes in many neighborhoods like yours have gone up in value.
You can use the money for anything it's yours. You can buy an investment property, payoff higher interest debt, or make home improvements. If you need twenty five thousand, fifty thousand or more, now is the time home values are up and so is your equity. We offer you a way to use it. No need to use your savings called New American Funding Now and see how much cash out you can get. Call eight hundred seven one oh three seven three nine, eight hundred seven one oh three seven three nine, eight
hundred seven one h three seven three nine. That's eight hundred seven one h thirty seven thirty nine nmls sixty six h six wwxs dot org. This is not an offer or commitment to lend subject to borrow. We're improperty qualifications, not all of borrow wres. We'll qualify terms and conditions apply equal housing express. Welcome back to Inside Analysis. Here's your host, Eric Kabanaugh. All right, folks, back here on Inside Analysis of fascinating conversation with Chris Gladwin
of Oscience and Christian Christman of sixth Sense Advisors. We're talking all about hyper scale, hyper scale of data warehousing. There's an ocean of data. Now, it's not a problem if you have the right approach. And we had a great question from one of our studio audience members who who mused, if we're never going to be able to see the raw data again, the question of analyzing behavioral anomalies or lineage and trustworthiness comes up pretty quickly, alike intrusion
detection or regulatory compliance, etc. And the break. You made a really good point, Chris, about how that the whether it's compliant or not, it's never going to be in just some field in a database. It's always an assessment based on multiple pieces of information. And to your point, at the scale of information flow. These days, it's going to be a complex layering of decisions based upon the data, right, go ahead without explanation.
Yeah, So if you, for example, you want to do intrusion detection, it's not like there's a data field somewhere in a database that says intrusion detection, yes or no. I mean you have to look at like a whole bunch of raw data and there might be a couple layers of analysis. You know, there's the analysis, then there's the analyzing. The analysis that says that is an intrusion and it characterizes that. So that's really what you're trying to get at. Is this meta meta information, you know, is
it an intrusion detection? Is this a risk? Is this trustworthy? You know, those are higher level things like you're building a credit score, you know for trustworthiness or business risk analysis. Well, that credit score might have a thousand different variables that go into it, and each of those variables might have you know, a million data points or about of the data points that
go and do it. So that's that's really more and more what you're seeing is you don't the raw data is not the thing that has business value. It's the analyzed result. Yeah, no, that that's exactly right. And you also made a really good point about just this new building block and the NVMe solid state drives and the interface and how it's it's literally the opposite of spinning disk. Can you explain that through audience, because that's a very compelling
story. Go ahead. I mean something we've we've asserted at ocient for the last ten years since we've been hearing about a real solid state drive. The parallel interface NVMe is the name of that non volatile memory express which is now, like I said, it's in your phone, it's in your laptop. It's it went from nowhere just a few years ago too. It's in everyone's
phone now. And that's also different because normally that kind of new technology, where it's new physics, new materials, starts and supercomputers and works its way down to consumer products, and here it just started in your phone, which is pretty amazing. And one of the things we think it's profound about it is we we assert that it's really the first new building block in computing.
And as we were talking about earlier hardware, at the end of the day is what dictates price performance and what happens and all you can do as a software architect and developer is make your software go as fast as the hard work and go you can go no faster than that. And the way that NVMME solid state works is basically, as you're just saying, Eric, exactly the
opposite on how a spinning disc work. A spinning disc is like a record player, and it has a ReadWrite head that sits in exactly one spot, exactly one time, and it sucks or spits bits onto that you onto or from that media. But it is in one place at one time, So physically what it wants to see is a serial stream of things to do,
a serial sequence one after another, because it's only in one place. Databases, when you're doing a count or a sort, or you know the things that databases do on the inside shuffle, you know, they actually want you to do a whole bunch of random parallel operations. But what the whole generation of multiple generations of databases data warehouses had to do was to figure out how to take what you want to be multiple parallel things and have them happen one
thing at a time on a spinning disk. Because anything in volume was on a spinning disk. Solid state is exactly the opposite. It doesn't want one thing to do at a time. The prior versions wanted two hundred and fifty six parallel tasks. Now they want five hundred and twelve headed to twenty four and so on parallel tasks. Her drive, you know, thousands and thousands
of times a second. It is exactly the opposite. So what that means in terms of your database, your software is what I'm describing is the interface between like a primitive database operation like a shuffle or sort, and the physical drive. That's called your IO layer. Okay, and if your IO layer has to treat your storage media exactly the opposite, what it means is you got to rewrite your whole IO layers. And that's forty percent of a database.
So if you really yeah, you could take an old architecture and set it on spinning disk, or sorry, take it off spinning disk, set it on solid state. It'll run ten times faster, but it should run ten thousand times faster. You're just leaving orders of magnitude. Now, you got to write your IO layer. But you'll find if you write your we write your IO layer, which we did anocent next thing, you know, you're gonna have to rewrite your memory allocators of Linux because they're not designed to
flow this much data. Next thing, you know, you're mucking down around with the actual assembler driver for that, you know, that mBMI solid state drive. And you also have to think called v tune, which is the thing that you use to tune the Intel CPU just right with the you know, the cashing layers. Like you're down there, like there's no there's no further down than where you are in the computing stack, and then you have to go up from there. So the next thing up from the IO layer
is you know, we call it the virtual machine. You know, it's the thing that once your parser kind of parses out your your query, you know, the parser will give you, you know, the parst query. Down to this virtual machine is optimizer and it's going to have to figure out
how to drop it down to the aisle layer. So that's the thing that figures out Every query can be done in infinite number of ways, So think about all the different ways and maybe think about it for half a second and then decide, all right, this is how we're doing it, and down you go to the aisle layer. So you know, that's another forty percent of your database. So you know, like parsers are not exactly rocket science
right now. You can go find a nice open source parser, but you end up having to rewrite your whole database stack, you know, down to the memory allocators if you really want this thing to go. And that's what we had to do with It took us, because it always takes at least five years. It took us five years. That's why you were saying version nineteen was our first production version, because it really was nineteen versions. But that's what you have to do. That's amazing. And you know, I'm
reminded of a couple of things. One I had doctor Michael Stonebreaker on this show and it was a long time. I was probably like ten twelve years ago, and at that point in time, he said eighty percent of the code that has written should just be thrown away. And what he was talking about is what you're talking about. That as the hardware layers change and as we get new building blocks to play around with, those old software applications were
designed to work in much different environments. And so how you designed this thing to you know, to do calls and to pull and do all polling or whatever it is that you're doing that has to be cognizant of the new stack, if you will, to your point with these NBMA, because it is completely different because it wants lots of things in parallel. You have to read write stuff, and it takes a lot of time to go all the way
down to that level. But unless you do that, you're missing huge chunks of potential optimization along the way, right Chris, Yeah, And you know I use the word assembler recently in that profit answer how often does that happen? In twenty twenty three? And you know that's just a little bit of code down there, you know, the Linux layer, but the database itself we had to rewrite and C plus plus and it's not like we don't know that that takes longer. It does take longer. But let me tell you.
If you have like two seconds of garbage collection and you're running a query where I don't know, you're gonna look at fifty trillion things and you want an answer in ten seconds, that means you're looking at five trillion things a second, your garbage collect for two seconds, and you got ten trillion things
to go put somewhere that's not going to work. So you know, next thing, you know, you're you know, you're down there and see plus plus and you're doing all this kind of crazy stuff where you're making everything stateless so you can just like parallelize the forever. You know, in our in our system it is. You know, if you've got a five hundred task per drive, I mean these are little solid state drives. You can put you know, twenty of them in a U. You can put twenty years
on a cluster. Next thing, you know, to get a million parallel tasks at every single layer of the stack, and things are flying. I mean you're doing five trillion things a second, you know, so every you know, it's flowing up and down that stack. Five trillion things a second, you know, million parallel tasks and each task there's a lot of things each second, you know. So it's just nuts. And so building something like that, it just takes forever. There's just no way around it.
Yeah, well, because you built two hyper scale. That's another interesting topic in and of itself. Right, if you're building for a particular end state, you're saying, I want this thing to good on this track, But no, you're saying I want this car to be able to go faster and faster on whatever track happens to come around, right, Isn't that the kind of difference in mindset. Yeah, we have a customer right now who we've done it on paper, but this is the first time we're doing it.
They want to demo with one hundred trillion row table, So we're building one of those right now. Yeah, I mean you design it. I mean, you know, we designed we actually one of the original customers that were still working with because it takes a long time to build something like this. They wanted something we're literally it was about a quadrillion quadrillion rows and a table, and they wanted to run queries that would hit most of that and give
answers in like ten seconds. And yeah, I mean you're going to do one hundred tillion things a second to answer that. So we've just been on paper, but designing on paper and actually having it run you know, in the physical universe or two different things. So we're kind of hitting in terms
of actual scale, like that hundred trillion scale right now. And we've got a number of customers that are like an exibite scale, you know, which will you get into the Quadralions. Yeah well, Folks podcast bonus segment is up next. Look these guys up online Ocean's Chris Gladman, Your Listenings and Hillistinea KCAA Lomolinda at one oh six point five FMK two ninety three CF Marino
Valley, NBC News Radio, I'm Chris Garagio. Officials in Hawaii put the confirmed death toll at ninety three from the Lahinah wildfire, and Maui Police Chief John Pelletier said search crews were using dogs to look for remains in the wreckage and that only three percent of the Destroit area has been covered so far. The Lahinah wildfires being called the deadliest US wildfire and over a century. A plane that crashed during an air show near Detroit was a Soviet era fighter jet.
The privately owned Mid twenty three was participating in the thunder Over Michigan air Show, one of the largest in the nation. Video posted on social media shows two people apparently ejecting from the plane before it crashed into an apartment complex parking lot forty miles east of Detroit. The two pilots were transferred to a nearby medical facility. They're in stay condition. No other injuries were reported.
Former Vice President Mike Pence says his former boss got bad advice from quote crackpot lawyers about the vice president's role in