What I'm Reading in the World of Data
The latest on universal databases, AI/ML-powered automation, data startups. Hadoop, and more
Hello again Cloud Database Report friends and a special shout-out to our new subscribers. Did you know that for only $5 a month, you get the benefits of a paid subscription, which include advance peeks at things I’m working on? It’s also a great way to support my work as an independent tech journalist and analyst. Thank you!
I took a few days recently to travel to the Silicon Valley to visit my son, who just completed finals for his MS degree in AI. I also did a couple swims in the very cold San Francisco Bay as I get ready for the Escape from Alcatraz Triathlon in early June. That’s Alcatraz with the lights on in the photo here.
So I had some time to catch up on my reading, and thought I’d share a few recommendations on interesting developments in our fast-moving world of data.
What I’m reading:
TileDB continues to grow its impressive list of investors, most recently adding Verizon Ventures. TileDB has developed a “universal database” that uses multi-dimensional arrays to support many different data types, and the company’s founder and CEO Stavros Papadopoulos is an iconoclast regarding the growing popularity of purpose-built databases. “We set out to debunk the myth around performance optimization in purpose-built data solutions,” he writes. “We proved that it is possible to achieve both universality and price-performance superiority over purpose-built solutions.”
Read the blog post: TileDB Secures Investment from Verizon Ventures
Many database providers have begun introducing AI/ML-powered monitoring and admin to automate the hands-on management typically done by DBAs. For example, Google Cloud’s new AlloyDB supports ML-assisted vacuum management, storage & memory management, etc. OtterTune, a startup with ties to Carnegie Mellon University, has developed an AI/ML-driven platform that automates database tuning for MySQL and PostgreSQL running on Amazon Aurora and Amazon RDS, with more to follow. As TechCrunch reports, OtterTune just raised Series A funding. Stay “tuned”!
Read the blog post: OtterTune, Which Taps AI to Optimize Databases, Raises $12M
Ben Stancil shares his thoughts on how to build a data startup. His advice: stop the hand-wringing about the many directions you could go, and just get on with it! “It doesn’t matter what you choose,” he writes. “If you’ve already done some research and narrowed a dozen mixed ideas down to a couple promising ones, both are probably good. The work you have left to do isn’t more research; it’s to make—and commit, fully and truly—to a decision.”
In his “Technically” blog, Justin sets up basic technical questions, then answers them in terms people can understand. Here, Justin tackles a subject for people who want to understand database environments: “What is a production database?” (Note: Like other Substack newsletters, some Technically posts are free, some paid. This one is paid but you can access it with a free 7-day trial.)
Finally, Gartner analyst Merv Adrian tackles the question, “What’s Next for Hadoop?” Based on client inquiries about Hadoop and other big data technologies, he says next steps are in sight. Not surprisingly, “data lake” and “lakehouse” are in vogue, but Hadoop is “far from dead.”
Read the blog post: The Hadoop Conversation is Now About What's Next