Automatic Indexing - Generally, It Depends

Dude … Where’s My Decade? If anything sums up my IT career’s arc since 2010, I’d say it’s that phrase. Somehow, I’ve managed to keep my head above water through the transition from HDD to SSD, from custom-built servers to Exadata Database Machine, and at least five different versions of the Oracle Database. Along the way, I’ve been fortunate to fulfill a personal goal of becoming an Oracle ACE Director for the the last half of that decade. That gave me the chance to meet IT folks on six different continents at dozens of Oracle conferences and OUG meetups.

Heck, I even got to chauffeur one of my heroes, Tom Kyte, to a Chicago Oracle User Group meeting!

As I enter my fifth decade in IT, I see 2020 as full of amazing opportunity, whether you’re a total Oracle newbie or a seasoned professional. Here’s where I think we’re headed:

Autonomy: The End of Tedium for “Helicopter DBAs”

Based on my deeper experimentation with both ADW and ATP this past year – especially with the new Autonomous Dedicated option that offers elimination of “noisy neighbors” – it’s apparent to me that the new role of Enterprise Data Architect can leverage that autonomy to further her career. With the announcement at Oracle OpenWorld 2019 of the upcoming launch of the Autonomous JSON database projected for early 2020, it sure looks like we’re going to be able to leverage key-value stores much more effectively than we have in the past.

The new Automatic Indexing features I’ve experimented with are just the tip of the spear. (You can download and then read more about my experiments in a recent white paper I published for Viscosity North America; if you are pressed for time, there’s a neat little video on that page too.) And I’m sure there are other autonomous features planned for later in 2020 that will finally free forward-looking DBAs from the tedium of normal day-to-day activities, especially just “keeping the lights on” for their databases.

A/I and Machine Learning: Nearing a New Singularity?

Based on the massive interest in AI and ML I’ve observed at conferences this past year, I think we’re nearing a new singularity in data-driven decision-making. I’ve spent some time recently experimenting with the Oracle Machine Learning (OML) features already built into Autonomous DB; if you’ve got a spare five minutes, be sure to take a look at this short video that shows how easy it is to leverage those features with just a little bit of SQL and a few mouse clicks.

The best thing about these ML toolsets is that they’re already built into the Oracle Database, so I don’t necessarily need to learn a completely new set of analytic functions or ML algorithms to take advantage of them. I’m going to continue my research in 2020 on several fronts, particularly in learning how to better visualize hidden patterns within my data with Oracle Analytic Cloud (OAC). If you’re interested, check out this short demo on how easy it is to leverage OAC for data visualization.

The Converged Oracle Database

Though it’s taken some time to get there, it appears we’re headed towards a new data-centric paradigm: Let’s get our business users exactly the data they need, from exactly where it’s currently stored – either intra-RDBMS, or extra-RDBMS! – exactly when they need it, with the utmost concern for its security. From the discussions I’ve had with folks at Oracle like Jenny Tsai-Smith, Oracle’s new Converged Database strategy is focused precisely on achieving these goals, so expect to hear and see a lot more about related offerings in 2020.

It’s Vegas, Baby!

Finally, if you want proof that the 2020s are going to be quite different from the past decade, I offer up this evidence: Oracle OpenWorld 2020 will be held for the first time, not in San Francisco, but in Las Vegas instead. It’s going to be quite the ride, my friends and colleagues! I look forward to exchanging knowledge with you all at upcoming OUG meetings, COLLABORATE20, KSCOPE20, and of course OOW20. Stay safe, have a great 2019 holiday season, and Happy Computing!

I returned from OOW18 energized and ready to enter a new phase of my IT career: Instead of fulfilling a pre-sales role, for the first time in the last four-plus years I’ll be concentrating on researching deeply the capabilities of Oracle Database technology, both in the Oracle Public Cloud and for on-premises solutions. And it looks to me like this transition is happening at the best possible time, based on what I encountered at OOW18 last week.

Front and Center: Autonomosity

Here’s my impressions of just two key takeaways from that conference:

Autonomous Database isn’t going away. Larry Ellison doubled-down on Autonomous DB at both of his keynote sessions. I wasn’t surprised by that because as Larry says, Autonomous Data Warehouse (ADW) and Autonomous Transaction Processing (ATP) are two of the more revolutionary offerings that Oracle has produced in some time. Yet in a way, they’re simply the logical progression of the “autonomosity” (to coin a phrase) of the Oracle Database and its powerhouse enterprise systems characterized most by the Exadata Database Machine.

I could definitely tell this development has quite a few of my Oracle DBA colleagues rattled, based on the standing-room-only crowds for just about any session whose title included the words the future of the DBA role. And maybe that’s good! IMHO, plenty of us have lost touch as to what being a DBA really entails; first and foremost, we should be focused on helping our customers – our development teams and business development units – build better applications and systems, and we can’t do that treating our role like it’s a job when it’s really a career – or, even better, a profession.

Automatic Indexing. Probably the most interesting new feature – coming in Oracle 19c, but entering beta testing as I write this! – is the ability for ATP to decide precisely which SQL statements need precisely which indexes to run better. Maria Colgan presented the basics of Automatic Indexing at our ACE Director briefing and I’m hoping to get a chance to poke it with a stick in the coming weeks. But here’s the essence of this new feature: it’s exactly what we’ve been doing as DBAs whenever a query or application malperformed – except now, it’s happening based on tireless observation and dispassionate decision-making through machine learning and artificial intelligence.

Stars On the Horizon, Rising In Prominence

And here are two topics that are definitely rising in prominence, based on interest I observed at OOW18:

Chatbots. The new technology that Ellison demonstrated in his second keynote when he submitted an expense receipt for reimbursement to the Oracle Digital Assistant was nothing short of amazing. Personally, I really enjoy doing expense reports more than most people – and I hate doing expense reports! – but the new chatbot-enabled technology made short work of submitting the request.

And that’s just a simple example of what chatbots promise. Ellison also demonstrated the ability of Oracle’s latest generation of Fusion Cloud applications to seamlessly create a mini-data warehouse with a few simple mouse clicks, and even answer verbal questions about Oracle’s ongoing efforts to hire the best and brightest candidates.

I even got a chance to experiment with the Pepper chatbot at The Hub for a few hours – one of eight new exhibits that my ACE colleagues and I manned to demonstrate the future boundaries of these new technologies. I was surprised at the flexibility of her communication interface, but not yet convinced Pepper will be taking my place as a speaker anytime soon.

Blockchain. Sessions on anything with the word blockchain in the title were heavily attended as well. That’s not a big surprise, of course; this technology, once it comes to full fruition, will offer some fascinating capabilities, especially the ability to preserve and traverse the myriad twists and turns of complex financial transactions. My colleague and fellow ACE Robert van Molken from the Netherlands – who just published a 500+ page book on Oracle Blockchain – discussed several alternative applications of blockchains, and I’m planning on diving deeply into this nascent technology in 2019.

Tag: Automatic Indexing

The Decade’s Almost Over. What Will 2020 Bring?