Learning Library

← Back to Library

Macro Trends Driving Data Lakehouse Adoption

Key Points

  • Three macro‑trends are driving analytics modernization: exploding data volumes and costs, evolving data consumption patterns (especially AI‑driven use cases), and a disruptive shift in data architecture.
  • Enterprises are spending significantly more—estimated ~30% YoY—not only on storing data across lakes, warehouses, and other stores but also on managing, governing, and securing the data lifecycle.
  • Business users and AI applications are demanding faster, broader access to data for tasks ranging from automated advertising optimization to human‑in‑the‑loop credit underwriting, intensifying the need for timely insights.
  • Growing regulatory and privacy requirements mean modern data platforms must embed robust security and governance to safely enable data sharing, while also delivering rapid “democratized” access for users.

Full Transcript

# Macro Trends Driving Data Lakehouse Adoption **Source:** [https://www.youtube.com/watch?v=N8TFwV4Q_w0](https://www.youtube.com/watch?v=N8TFwV4Q_w0) **Duration:** 00:07:58 ## Summary - Three macro‑trends are driving analytics modernization: exploding data volumes and costs, evolving data consumption patterns (especially AI‑driven use cases), and a disruptive shift in data architecture. - Enterprises are spending significantly more—estimated ~30% YoY—not only on storing data across lakes, warehouses, and other stores but also on managing, governing, and securing the data lifecycle. - Business users and AI applications are demanding faster, broader access to data for tasks ranging from automated advertising optimization to human‑in‑the‑loop credit underwriting, intensifying the need for timely insights. - Growing regulatory and privacy requirements mean modern data platforms must embed robust security and governance to safely enable data sharing, while also delivering rapid “democratized” access for users. ## Sections - [00:00:00](https://www.youtube.com/watch?v=N8TFwV4Q_w0&t=0s) **Untitled Section** - - [00:03:07](https://www.youtube.com/watch?v=N8TFwV4Q_w0&t=187s) **Data Governance and Architecture Evolution** - The speaker highlights how rising regulatory demands and the push for rapid, democratized data access drive enterprises to adopt secure, governed data practices and flexible, cost‑optimized cloud storage architectures. - [00:06:26](https://www.youtube.com/watch?v=N8TFwV4Q_w0&t=386s) **Lakehouse as Open, Cost‑Saving Highway** - The speaker describes how an open data lakehouse eliminates vendor lock‑in, cuts warehousing costs by up to half, and uses a highway/toll‑road analogy to illustrate flexible, low‑cost analytics. ## Full Transcript
0:00In a previous video, we talked about the 0:02data lakehouse concept and shared a 0:04story about how data lakeous are much 0:07like the operations of a commercial 0:09kitchen in a restaurant. So, definitely 0:11check out that video if you haven't seen 0:12it. Today, I'd like to discuss more 0:15about the key drivers for and values 0:19delivered by an open data lakehouse 0:20architecture as well as share a couple 0:23examples. And to help me do that, I'm 0:25very excited to invite Edward Calvas, 0:27director of product management for IBM 0:29databases to join us. Edward, thanks for 0:31being here. 0:32Hey, love. 0:33So Edward, let's start with the major 0:36macro trends that we're seeing and that 0:38are leading organizations to modernize 0:40their analytics infrastructures. How has 0:43the use of data shifted drastically in 0:45the past couple years? 0:47Well, of there are three major macro 0:49trends that we're seeing in the market. 0:51First, the amount and cost of data is 0:55exploding. 0:57Second, data consumption patterns are 1:00expanding and changing. 1:03And third, data architecture is being 1:05disrupted and transformed. 1:10So, could you speak briefly about each 1:12of these? 1:13Sure thing. 1:15So there's no doubt that the amount 1:18of data is expanding rapidly, but also 1:22it's coming from a variety of different 1:25sources and in all sorts of new data 1:30formats. 1:34What this means is that to manage all of 1:37this data, enterprises are spending more 1:40money and some estimate that that is in 1:42the range of 30% year-over-year. 1:46Okay. So, when you talk about the cost 1:48of data, are you just referring to the 1:50cost of storing it in different 1:51repositories like data lakes, 1:53warehouses, or other stores or are you 1:56also referring to the cost of managing 1:58and governing the life cycle of that 2:00data? 2:00Well, it's actually both. Let's talk 2:02about the patterns of data consumption. 2:05There's an everinccreasing demand for 2:07the use of data, especially from 2:09business users. There's no doubt that 2:11analytics 2:16has become an essential component of 2:18almost every job and certainly AI, the 2:22use of AI is expanding rapidly. 2:25Now, this doesn't mean that every 2:26business user needs to become an AI 2:28expert, but it does mean that more and 2:30more we're seeing AI being used to 2:32automate and optimize certain decisions 2:34at scale, such as advertising campaigns 2:37or supply chain networks. 2:40AI is also being used to augment human 2:42in the loop decision-m such as credit 2:46risk underwriting. 2:48This means enterprises are always 2:50looking for more data 2:54and use it to drive new insights, 2:58right? And and you know what about um 3:01the data privacy and data regulatory 3:03concerns that are around AI? 3:05Sure. When you combine this with 3:07increasing regulatory standards, 3:09enterprises will require higher levels 3:11of built-in data security and governance 3:14in order to enable this data sharing and 3:16consumption. Absolutely. And you know 3:18another thing that we hear a lot is the 3:20democratization of data. So it's about 3:23create it's about time to value right. 3:25uh business users, yes, they need data, 3:27but they need it like yesterday, right? 3:29It doesn't do the user or the 3:31organization a lot of benefit. If it 3:34takes long complicated processes for 3:36users to get access to that data so to 3:39get the most value out of the data, it's 3:41got to be consumed as quickly as 3:42possible. Um and of all while still 3:45adhering to those governance and 3:47compliance policies. Would you agree 3:48with that? 3:49Absolutely. That leads us to our third 3:51point, which is architecture. 3:54Organizations are realizing that the way 3:56the data is managed needs to change. The 3:59emergence of commodity cloud object 4:02storage 4:04and the adoption of open data formats 4:08is really allowing enterprises to 4:10increase the return on investment uh in 4:12data management. And they're doing this 4:15through the optimization of the price 4:18performance 4:20of their workloads across different 4:22storage and compute tiers. Now what does 4:24that mean? It means that organizations 4:27can benefit from having the right tool 4:30for the right job at the right cost 4:32instead of defaulting into a data 4:35warehouse which may be appropriate in 4:38some cases but can also become very 4:39expensive and ineffective in others. 4:42Right? So what I'm hearing is more data, 4:45more users and more uses of that data, 4:48right? And all while still uh better 4:51ways to share and manage access around 4:54it, right? Um so Edward, how are these 4:58aspects related to the key values that 5:00are delivered by an open data lakehouse 5:02architecture? 5:04Great question, love. We see three key 5:06values that are delivered by an open 5:08data lakehouse architecture. First, an 5:11open data lakehouse provides the 5:13foundation for users to easily and 5:15cost-effectively access, store, manage, 5:18and unify large amounts of data and from 5:22different sources and in different 5:23formats. 5:25Second, an open data lakehouse can be 5:28easy to deploy within existing 5:29environments, providing users fast 5:32access to more data without having long 5:34procurement, onboarding, or data 5:36pipelining and wrangling processes, 5:38making it much easier to consume. And 5:41third, open data lakeouses can optimize 5:43your analytics workloads to run where 5:46they perform the best and are most 5:48costefficient. All as part of an 5:50integrated architecture. 5:54Wow. So, you know, I've noticed one 5:56thing. You've said the word open a lot. 5:58Can you explain to us what the 6:00difference is there? 6:01Sure. I'm glad you brought that up. A 6:04lakehouse should leverage the 6:05capabilities across existing data and 6:07analytics environments. If you already 6:10have data and analytics workloads in a 6:12data warehouse or in a Hadoop data 6:14lakeink, that's okay. You shouldn't be 6:16forced to migrate or rip and replace 6:18that environment in order to get started 6:20with the lakehouse. But what about new 6:22data and new workloads? 6:24Well, a lakehouse should be the starting 6:26point for new data and new workloads as 6:28well as provide a modernization path for 6:31existing environments over time. 6:33You know, the other thing that we hear 6:34uh often is customers getting stuck with 6:37one vendor. 6:37Well, al open also means that you're 6:40always in control of your data and you 6:42aren't forced into proprietary data 6:44formats or specialized tooling in order 6:46to use it. It also means that you can 6:48maximize the use of your data without 6:50having to make copies of it and move it 6:51around. And this provides for lower 6:54costs, higher productivity and better 6:57governance, which ultimately leads to 6:58what enterprises are looking for, which 7:00is more trusted decisions. 7:02Wow. So Edward, what I'm hearing is a 7:05data lakehouse 7:07is like a network of highways. 7:11Some of them have tolls on them and some 7:14of them don't like regular freeways. And 7:17a lakehouse allows you to go on the toll 7:20road, pay when you need to get somewhere 7:22really fast, but when you're not in a 7:23hurry and there's no traffic, take the 7:26regular highway. Love, that's a great 7:28analogy. Think about using a data 7:30lakehouse to cut your data warehousing 7:32cost by up to half by optimizing the 7:34price performance of your analytic 7:36workloads. or like you're saying, saving 7:38money by driving on the freeway while 7:40there's no traffic and you're not in a 7:41hurry. 7:42Wow, Edward, that sounds really 7:44exciting. Let's hit the road. 7:45Let's do it. 7:47Thank you. If you like this video and 7:49want to see more like it, please like 7:51and subscribe. If you have questions, 7:53please drop them in the comments below.