A series of wise decisions to build an open cloud ecosystem have propelled the cloud giant to prominence. However, converting that openness into a sustainable advantage will be difficult
But Google Cloud has never taken the same approach as its hyperscaler competitors. Google was into open source long before it was cool, and it was also among the first to form partnerships to meet the needs of its customers. Sure, not every aspect of Google's warm-and-fuzzy approach has worked right away; at times, the company has needed to become a little more "boring" in order to reach enterprise buyers. Still, Google's honest efforts to use openness as a strategic way to compete and work together are paying off, and this week's Google Next 2022 event will be the climax of those efforts.
Gerrit Kazmaier, vice president and general manager of data analytics at Google Cloud, stressed the importance of "open" (open source, open standards, open data) throughout our interview, mentioning it 29 times (yes, I counted). Add this to the 100+ times the word "open" was used in the blog posts, press releases, and other materials released by Google Cloud today, and Google's overall message is clear:
Google wishes to be known as the "open cloud." This sounds superfluous, but it could be useful in practice. Not to mention that it's very hard to do and needs a new way of thinking about who owns the product and the data.
Making open pay a possibility
As I previously stated, Google's embrace of open source enabled it to gain a strong foothold against AWS, which was the first to enter the cloud market. In the case of Google, I suggested that, while open source contributor counts do not guarantee success, they can play a role in long-term, customer-focused strategies and help reshape markets. That's the wager, and it appears to be working.
Following that, Google Cloud went further to establish itself as the "most open data cloud ecosystem" by "unifying all data, from all sources, across any platform." The words "all" and "any" imply some exaggeration, but consider how the next announcements have brought Google Cloud to the point where it can make that claim:
Given that Databricks developed Delta Lake and that Databricks and Google Cloud compete for data warehousing workloads, I asked David Meyer, senior vice president of product management at Databricks, about the Delta Lake integration. Meyer says it all comes down to customers. "Our customers advised us that we should be on Google." Why? Larger Fortune 1000 companies, on the other hand, "need some diversity in cloud from a leverage perspective, but also from a data estate perspective," Meyer says. These businesses already use Google Ads, so adding Google Cloud as they expand their cloud footprints makes a lot of sense. This will be made easier if they can store their data in Delta Lake. Customers can apply BigQuery to data in their Delta data lakes without having to move it thanks to this partnership and Google Cloud's support for the Delta Lake format.
Win, win, win
Google also announced some housekeeping changes (such as rebranding all of its business intelligence services as Looker), but the company went above and beyond by deeply integrating Looker and Google Workspace to make BI-driven insights available in the familiar productivity tools (Google Sheets) that customers will use on a daily basis. This isn't "open" in the sense of open source, but it is open in terms of lowering barriers to data use. Other clouds have accomplished this by making it easier to use, such as MySQL or Linux. Google Cloud provides similar services but goes one step further by making data easier to use rather than just data infrastructure. Google's Vertex AI Vision is similar in that it makes computer vision and image recognition AI, which can be hard to use, easier for data practitioners to use.
Google may have a slew of PhDs on staff, but thanks to new initiatives like these, you may not have to. This is good because, as Kazmaier pointed out, enterprise data estates will only become more complex.
Data is being made available everywhere
Companies may claim to be "all in" on a single cloud, but the messy reality is that they aren't, or rarely are. CIOs can play Whac-A-Mole with application creep across multiple clouds, including on-premises infrastructure, but "data is spread across multiple clouds for the vast majority of companies," according to Kazmaier. As a result, multicloud does not imply deploying the same solution across multiple clouds and having independent silos of the same technology. Kazmaier says, "It's about connecting the data from different clouds into a whole data landscape."
This is the vision driving Google Cloud's embrace of multicloud, which is supported by Anthos and other technologies. It's also why, as of this week, BigQuery can now analyze unstructured streaming data, allowing enterprises to combine structured and unstructured data analysis in one place (called BigLake). That is a massive, extremely difficult problem to solve. Google isn't announcing a "data cloud," but rather "a data warehouse that pulls data from the cloud." No, it means that operational data from databases like MongoDB can be easily analyzed using data warehousing/analytics services and AI/ML activation systems.
It's a lofty goal. It's quite impressive. However, it is extremely difficult to achieve in practice because it requires Google to think beyond itself when developing customer-centric products.
The only way to make it work is to stop thinking in terms of total control over the customer experience and data. Nobody denies that Google has created some of the industry's most innovative open source code: Kubernetes, TensorFlow, and so on. But, I'd argue, what Google Cloud, and each of the clouds, needs to do now is not be the originator of everything. The cloud is too large for any single vendor, regardless of size. No hyperscaler is ever "hyper" enough to create solutions for every need.
So far, Google appears to concur
As previously stated, the company has always prioritized partnerships, but at this year's next event, the company has added more substance. According to Kazmaier, Google Cloud must "have 100% open APIs" to enable a deeply integrated partner ecosystem, but this is insufficient. "This also implies that the APIs we use in our first-party products are the same APIs we expose to our partners." Yes, there will be exceptions to this rule. Many workloads, according to Kazmaier, are "best served by a partner, and our strategy is to open up our APIs so they can build." We don't see ourselves as competitors to them. "
If this appears to be a novel approach, not just in the cloud but in enterprise computing in general, it is. However, it is consistent with Google's core principles. Perhaps it is the logical conclusion if we begin with the realization that data continues to explode in all directions. If we assume that data will grow across clouds, that it will be both structured and unstructured, that it will necessitate both real-time and batch-oriented analysis, and that it will be complex in a variety of other ways, then Google Cloud's approach becomes unavoidable. "An open platform will be the best choice for customers because it will ultimately offer them the greatest flexibility, the greatest degree of choice among multiple solutions, and the shortest time to value than anything else," Kazmaier concludes. That logic is difficult to refute.