By David Baum
Today’s AI systems bring a high degree of automation to everyday activities, from selecting the fastest driving routes to getting to know your music preferences. The most-advanced AI systems use machine learning algorithms to analyze current conditions and learn from experience. As these self-aware systems get more capable, they are transforming the IT industry as well—including Oracle Database, the world’s most popular enterprise data management system.
“With Oracle Autonomous Database Cloud, all management tasks are fully automated, including all database tuning chores,” says Çetin Özbütün, Oracle senior vice president. “Your data is automatically compressed and encrypted, and your queries and transactions run fast, because our autonomous database service is built on Oracle Exadata and Oracle Database, which represent 10,000 person-years of engineering effort.”
Oracle Autonomous Database Cloud currently includes two services: Oracle Autonomous Data Warehouse is designed for data warehouse and analytics workloads, while Oracle Autonomous Transaction Processing enables in-memory OLTP access.
Fueling Administrative Efficiency
Like all other Oracle Cloud services, Oracle Autonomous Database Cloud services scale to fit your capacity requirements and are delivered via a pay-per-use model—which Özbütün says can cut runtime costs by as much as 90%. These capabilities are especially attractive to organizations such as Drop Tank that want to eliminate mundane database management tasks by standardizing on a self-driving, self-securing, self- repairing database.
Drop Tank, a leading loyalty technology and rewards company, is using Oracle Autonomous Data Warehouse to create new loyalty solutions for gas station operators. The company partners with loyalty programs to help them add fuel purchases to their overall member experience and then works with service stations and their attached convenience stores to present compelling offers to consumers.
Oracle Autonomous Data Warehouse allows us to focus on analyzing the data rather than on managing systems. We don’t want to have to deal with management, patching, and security—and we can’t afford to have our systems down for maintenance.”—Tim Miller, Vice President of Technology, Drop Tank
In addition to creating effective loyalty solutions, Drop Tank mines insights from point-of-sale data to help gas station operators and consumer packaged goods (CPG) companies identify trends and make intelligent decisions about future promotions and sales strategies. The data helps these partner companies understand and influence consumer purchase behavior at thousands of service stations and convenience stores. Oracle Cloud services, anchored by Oracle Autonomous Data Warehouse, make it possible.
“By collecting point-of-sale data, we can discover trends and figure out what offers are working, and where,” says Tim Miller, vice president of technology at Drop Tank. “More and more, the data is driving our marketing and promotional calendars. We use analytics to get the right offers to the right people at the right time.”
Previously, Drop Tank used on-premises databases and management software for these tasks, but as the company grew, maintaining this complex IT environment became progressively more difficult. “We had to administer the virtual machines, patch the software, manage the application servers, and tune the databases, which meant we either had to hire system administrators for each technology component or become experts ourselves,” Miller recalls. “We’re not a large company, and as we grow, we don’t want to have to hire a team of database administrators to manage databases, update software, create tables and indexes—all of the things DBAs historically do. Oracle Autonomous Data Warehouse allows us to focus on analyzing the data rather than on managing systems. We don’t want to have to deal with management, patching, and security—and we can’t afford to have our systems down for maintenance.”
CPG companies typically sell through distributors that place goods at retail stores, but the consumer experience is a “black hole” because the point-of-sale data rarely reaches the manufacturer. Many of these firms don’t even know how much their products sell for. Drop Tank can capture the point-of-sale data and share it with the pertinent companies to unlock these customer insights. Everybody in this interlinked value chain wins: Gas stations can boost sales and drive more repeat business through greater customer loyalty, CPG companies can produce more of the items that sell well and yield the best margins, and consumers are happy to find their favorite products on the shelves.
“We’re seeing a future that involves a lot more transactions, a lot more data, and a lot more need to make use of that data in order for us to be efficient,” reports David VanWiggeren, CEO of Drop Tank. “Universal product codes, quantities, prices, taxes—all of those offers and items can now be related to individual consumers.”
According to VanWiggeren, service stations love Drop Tank’s loyalty programs because they “drive incremental purchases”—and not just gallons of gasoline, but convenience store items as well. Drop Tank’s largest retail network includes Marathon Petroleum, whose gasoline is sold through approximately 5,600 independently owned retail outlets across 20 states and the District of Columbia. Given the size of the network, other national loyalty programs are now using Drop Tank to engage with their members each time members fill up or make purchases in the convenience store.
The Value of an Integrated Cloud Platform
Drop Tank has found it easy to extend the capabilities of Oracle Autonomous Data Warehouse with other complementary Oracle Cloud services. For example, Drop Tank uses Oracle SOA Cloud Service to gather point-of-sale data and upload it into Oracle Autonomous Data Warehouse, where it is tied together to drive analytic insights. In the future, Drop Tank plans to use Oracle Analytics Cloud to analyze and visualize the data—both at the store level to track sales and in aggregate to help its partners understand short- and long-term purchasing trends.
Drop Tank also plans to use Oracle Autonomous Transaction Processing to execute a growing volume of point-of-sale transactions—from approximately 5,000 transactions per day now to an estimated 100,000 transactions per day by the end of 2019. “We anticipate a huge increase in memberships and transactions,” Miller says. “Oracle lets us automatically scale its cloud services to handle increasing volume. It’s easy to add capacity during big promotional campaigns and then scale back down again. If we aren’t using it, we aren’t paying for it.”
Although the premise of “pay only for what you use” has become an anthem for the cloud computing industry, Miller says that, in practice, not all cloud services are easy to scale. “With some vendors, you have to shut everything down and rebuild your information systems,” he explains. “You can’t just say, I need another node. With Oracle Cloud, on the other hand, we can spin up new nodes and within five minutes everything will be ready to go.”
Unlimited Scalability and Exceptional Performance
Having instant scalability is equally important to Aker BP, one of the largest independent oil companies in Europe, with full-fledged exploration and production operations concentrated on the Norwegian continental shelf. Previously, if Aker BP needed a new data warehouse, it could take weeks to provision the infrastructure, install the software, create the database tables, load the data, and tune the system for the workload at hand. Now, using Oracle Autonomous Data Warehouse, the company can implement a new data warehouse in minutes—and the database tunes itself to ensure maximum performance.
“The whole process of using Oracle Autonomous Database Cloud is very easy, from setting up an account to creating an instance to loading data,” says Erik Dvergsnes, senior architect at Aker BP.
According to Dvergsnes, creating a new data warehouse with Oracle Autonomous Database Cloud required four simple steps: naming the database, specifying the number of CPUs, determining how much storage was needed, and entering a password. After that, he was able to connect the new cloud database with Oracle SQL Developer and immediately start running queries. “I copied some Data Pump .dmp files to the cloud and ran some Data Pump imports on those files,” he explains. “The process is clearly documented on Oracle’s website. I could immediately start querying and playing with the data.”
Dvergsnes says the performance of Oracle Autonomous Data Warehouse is exceptional because of the way it automatically handles storage cells and indexes. As a result, Aker BP reduced report runtimes from 20 minutes to less than 1 minute for a 1 TB data warehouse with 1.2 billion rows, without any manual tuning by the IT staff. “It’s like having the world’s best DBA working for you,” he adds.
In addition, thanks to the Hybrid Columnar Compression feature of Oracle Database, Aker BP was able to reduce its data storage requirements by more than 70%. For example, one database went from 900 GB to 240 GB, once it was compressed.
Partly as a result of these performance and efficiency gains with Oracle Autonomous Database Cloud, Aker BP is now looking into decommissioning one of its Oracle Exadata systems and transferring the associated database licenses to Oracle Cloud as part of the Oracle Bring Your Own License program. “I’m really excited after testing Oracle Autonomous Database Cloud,” Dvergsnes states. “My focus is to migrate more and more databases to the cloud, with the goal of switching off our Oracle Exadata by the end of 2019.”
A Self-Scaling Database That Eliminates Downtime
Aker BP runs a 24/7 oil production and drilling operation, which means its information systems for issuing work orders and monitoring projects can’t tolerate downtime. These round-the-clock operations have always made it difficult to schedule database maintenance, especially since work schedules on the rigs can be interrupted at any time due to rough weather in the North Sea. The self-patching capabilities of Oracle Autonomous Data Warehouse could alleviate the need to ever shut down this important database for maintenance. “Many of our systems are mission-critical, so planning downtime for software patches and security updates is extremely difficult,” Dvergsnes says.
It’s like having the world’s best DBA working for you.”—Erik Dvergsnes, Senior Architect, Aker BP
While the self-patching capabilities will eliminate downtime for essential reporting activities, having capacity on demand will reduce the need to maintain excessive capacity to accommodate peak analytic loads. For example, Aker BP must periodically generate reports about the status of its offshore equipment, which involves investigating a constant barrage of alarms and alerts. The IT team sometimes has to review months of alert data to trace issues with a particular piece of gear, such as an oil pump. Which systems were involved? What alarms were generated? Who followed up, and what actions were taken?
“We might have to submit a query on a specific tag on a specific time stamp to find out exactly what types of alarms were generated from a specific incident,” Dvergsnes explains. “Due to legal reasons, we need to be able to always query this data.”
These report requests are sporadic—sometimes daily, other times monthly—and the associated queries can consume massive compute and storage resources. “It’s a perfect candidate for Oracle Autonomous Data Warehouse, because we can turn off the database when it’s not in use and then scale it back up again when we need it,” Dvergsnes adds. “It’s critical to produce these reports, both for the government and for our partners. If the reports are delayed, Aker BP has to pay big fines. If a deadline is looming, we can simply scale up to add CPUs and get the reports done in time. We can spin down a CPU to zero if nobody is using it. Oracle Autonomous Database Cloud is outstanding in the market, as far as I can see.”
Enterprise-Grade Technology for Smaller Firms
One of the driving forces behind Oracle’s new autonomous services is to enable small and midsize organizations to benefit from enterprise-grade technology that was formerly available only to large companies with many IT professionals on staff. Quality Metrics Partners (QMP), a healthcare holdings company specializing in ancillary service management, is a case in point. QMP used Oracle Autonomous Data Warehouse to create a cloud-based healthcare technology platform called CAREiQ that enables the company to improve its internal processes and offer cloud-based diagnostics, billing analysis, and other lab services to healthcare providers. Rather than allotting months to creating an on-premises analytics platform, QMP’s clients can typically implement CAREiQ within a week. Once they’re on the system, they can get diagnostic screening results within hours—a cycle that formerly took days.
“Perhaps the most important outcome of our new cloud service is better care for the patient population,” says QMP CEO Michael Morales. “Oracle tools enable us to educate physicians and help them be more aware. They can discover patterns in the data that assist with their diagnosis and detect potentially overlooked disease markers. They can also receive alerts, so they can be proactive about conditions that they may not have caught otherwise.”
QMP began as an owner of diagnostics laboratories and now delivers data analytics and other services to hospitals, clinics, and laboratories. Its analytics systems pull client data from various sources, including multiple types of electronic medical records systems, laboratory information systems, and billing systems. It uses Oracle Data Integrator Cloud Service to gather the information and prepare the data. Oracle Data Integrator Cloud Service reformats transactions, combines data from multiple tables, and then loads the data into Oracle Autonomous Data Warehouse.
Having a fully managed data warehouse service helps QMP eliminate manual, error-prone management processes. “We don’t need to have a DBA constantly managing the system,” says Steve Chamberlin, CIO of QMP. “We can grab data from another database, another cloud service, or lab files and then load it directly into the warehouse. Oracle establishes the architecture and manages the process. It’s actually more efficient than what our DBAs constructed.”
The new platform enabled QMP to avoid hiring a projected 15 to 20 additional analysts per laboratory to manage its growing data, saving the company hundreds of thousands of dollars in employee costs. Routine management tasks such as provisioning, patching, performing backups, tuning, and applying security patches have been automated by Oracle’s autonomous cloud service. “We help other healthcare organizations leverage their data so they can provide better service to their patients and prosper in this competitive industry,” Morales explains. “We can tap into the database of a clinic or hospital and pull data into Oracle Autonomous Data Warehouse. It loads and scales immediately to the required size.”
Built-in adaptive machine learning technology automatically tunes, upgrades, and patches the database while it’s in operation, even as workloads increase and decrease. “We simply create a new instance in the cloud, and we can load a new client’s data immediately,” Morales continues. “Performance and scalability tuning take place automatically behind the scenes. If you throw a bigger workload at it, the system accommodates it for you.”
We simply create a new instance in the cloud, and we can load a new client’s data immediately. Performance and scalability tuning take place automatically behind the scenes. If you throw a bigger workload at it, the system accommodates it for you.”— Michael Morales, CEO, Quality Metrics Partners
According to Chamberlin, Oracle’s new database also helps QMP comply with the Health Insurance Portability and Accountability Act (HIPAA) regulations governing the security and privacy of patient data. “Oracle provides a secure environment that uses encryption technology and other advanced database security technologies to help us with the HIPAA regulations,” he adds. “In addition, the Oracle data centers are audited every quarter. That’s way more often than what most small companies can afford to do.”
Looking ahead, Morales hopes to use Oracle Autonomous Transaction Processing as well, saying the new OLTP service will represent a “big win” for everyone involved: QMP can improve operational efficiency and expand its business model; doctors and clinicians can make better, more-knowledgeable decisions; and patients will get test results and diagnoses more quickly than ever before.
“We help doctors make intelligent decisions and focus on what they’re good at,” Morales concludes. “Autonomous technology from Oracle enables us to provide them with simple tools that can help them run their practices better so they can more effectively manage the entire patient population under their care.”
LEARN more about Oracle Autonomous Database Cloud.
Photography by Paul S. Howell and Ton Hendriks
Illustration by Pedro Murteira