Speaker 3
I think this case back to the point like where ben was talking about the model that in that morav controller relationship. And i think the model for data is more complicated sinside that web backen, you're just talking about a relational data base that's abstracted, or no sequal data base. Whereas for what we have to do in data, we have to build these really complex layers of like entities and metrics and semantics that then are usable. And i think that's where the model is deeper in data than it is in sufcoengineri in many instances. But ye, can can more happen inside that that model? Yes, i think so. And even if it's just kind of to the left and to the right of the dag, if, you know, where things could happen before and things could happen afterwards, but it still fits within it, that still, just that alone is hugely powerful compared to what we have to day. And you know, you've got to like, fowl entering that space, unabling that space, tha, that wo quite excitid.
Speaker 1
So i have laid three points on this, and one's a question. One's a question to you, jriston the first question is, well, tit's out of the question, i guss. The first thing is, daybricks, to me, feels like the like, biggest, forty billion dollars astake. Oh, well, were i not mid nigt like they they did greak. They're obviously great. They're ought to be rich. But it feels like they try to market this thing is like this complicated piece of technology that nobody quite kits and the way that i wants to get data bricks is it's a tool that someone smarter than me uses. And like everybody i know seems to have that same thing where it's like, i don't know hw to use data bricks. But people who are smarter than be sartainly do they seem like its power and snowflate base. They seem to come along and say, well, we built data bricks, but sust loks like a sequal data base everybody can use that. The question that i have anata is like, and this compute thing, is there anything stopping these companies from essentially just saying, hate we store all heur data in one basically nes thre. We put a bunch of differen engines on top, snow flakes, and initial engine is just like a sequel thing that looks like post press cratin. We have a pathon engine, like rather than this kind of, oh, it's a spark integration. It does all these sorts of things, that's like this kind of confusing, monolithic enterprise piecis of square. To me, what i really want is just, like, all my data lis in one place. I can connect to it different compute engines that speak different languages. And right now, that's kind of what like data bricks does. It's kind of what spark does, but in this way thati feels very hard to get your head aroundand and so i think like we actually could just solve this by sending, hey, actually. And i suspect snowflake will do this with, like, we should hae a way to connect it with python. It just looks like python, but underneath it, all of your data is there. And i successful, the question i have for you, trust in, and this is like the wild idea, how far clark can you pull compute and storage, like, can those be different companies? And is there anything that actually long term stops d b t from being a compute layer that sits on top of s three and cuts out snowflake out of this entire oh
Speaker 2
gosh, youwel you. You opened all the cans of worms all at the same time, which is something that you're famous for doing. Letme pick up the thread on thead. So they answer to like cn, d, b, t, d x. I think our preferences to do as as little as possible when it comes to computin storage, as little as possible, limit zero. And i spoke at the sub surface conference recently, and was on a panel with ryan blue from iceberg. And iceberg, and this is towards the edges of my deeply technical knowledge, but iceberg is a table format. It's on a file format. Parquet is a file form. But iceberg is a way to organize a series of parquet or other files in cloud storage and have a unified, lke medidata e, way to figure out what file to go to when you run a certain query. And so that, that kind of feels like a big part of what a data base does. But the interesting thing is, it actually doesn't have a sequel or other end point to connect to do processing. It doesn't have the compute layer. It's 's just the table layer. And so dremio, the host of subsurface is, loves this because they are a compute layer that doesn't have, they don't have a strong opinion about how you should be storing data. And, you know, we've talked about the stuff as data lakes historically, but the interesting thing about the historical data, like paradime, is that it's been a compute layer pared with a chiton of parquet files. And, ok, maybe you'velike, got some other maybe you've got ike hive table for men ind there, so something to do. But i think that we're actually getting better at this table level, which i think is the right abstraction to pare the storage and compute with. So, and you've also got snowflake is supporting iceberg. I think others are porting icebergs. So you could start to imagine ban you were just saying, like, i've got all my data in this one platform, and i can access it via like, multiple different engines. And i think that that is like, a, that is certainly an approach, and, and my guess is that all the data platforms will want to move towards that world. But i think that there is this other approach of, like, all my data is stored in iceberg tables, and i can have two contracts. I can have a snolay contract and a data brick contract, and actually each one of them is a little bit betterat, like, the thing that they are best at, and they're both reading and writing the same set of files. Now, ok, there's like, some, some like, i kind of, like, rounded off some corners there, and i realize that. But i think that there is a version of the world that looks kind of like that. So
Speaker 1
your saying is, we need to unbundle tables, which is really going down to level of this that's like, oh, my god, worg. It ahead like files in one place, and like the mack to the files on another is is a new level of unbundling of this. Holes
Speaker 2
you wat you weren't ready for them. Tas
Speaker 1
iam, not prepared for that. I e cato make sense. I hink hers like some clever stuff on ter tha's each got a cool but like that, i don't have the mental fortitude of that one.