Today, data is growing exponentially. At this rate, it is fast outpacing manual data management. It is becoming increasingly difficult to process data efficiently and securely in order to deliver business-critical insights. Moreover, even in this age of digitization across industries, the data of many companies is found in separate silos. This is a major impediment in extracting the real value of data and it also leads to duplicated efforts. Such a scenario often results in different business units of a company seeking similar insights from the same data. With the coming of cloud technologies, such problems can get even more magnified. Such problems can be rectified through the use of an autonomous database which is self-driving, self-securing and self-repairing. This is the power that the Autonomous Database gives to businesses, claims Tirthankar Lahiri, SVP – Data and In-Memory Technologies, Oracle.
What is an Autonomous Database?
An autonomous database is a cloud database which utilizes technologies such as Machine Learning (ML) to automate database tuning, backups, security, and updates, along with other routine data management tasks traditionally performed by DBAs. However, unlike the traditional database, an autonomous database accomplishes these jobs without human intervention.
“Autonomous Database has leading Oracle Database software, runs on leading database infrastructure Exadata, which is our engineered system and which runs on Gen 2 Cloud by Oracle. So, all of Oracle’s capabilities come together,” says Lahiri.
The company has opened enterprise-grade Gen 2 Cloud region for customers in India. The Gen 2 supports legacy workloads even as it delivers modern cloud development tools.
Oracle’s Autonomous Database Leverages Machine Learning Technology
According to Lahiri, Oracle’s Autonomous Database is the world’s first and it redefines database management by using ML and automation to eliminate human labor, errors, and manual tuning; this reduces cost and complexity and ensures higher reliability, security, and operational efficiency. Autonomous Database supports a complex mix of high-performance transactions, including reporting, batch, Internet of Things (IoT), and machine learning in a single database. This simplifies application development and deployment and enables real-time analytics, personalization, and fraud detection, adds Lahiri.
“Autonomous capabilities in our cloud is a very special thing. It is the best of breed and best in class – be it on the database side, infrastructure side, networking, administration, and everything overall. We believe Oracle Autonomous Database is bringing to you the future of data management. We don’t believe a retailer, or a mid to small sized bank, or in that case even a large bank, really wants to be in the business of managing a database on their own. They want to be in the business of innovation, to grow their core business,” points out Lahiri.
The Autonomous Database comes with a whole bunch of native AI and ML capabilities and has a lot of different models. The company is trying to make it accessible, so that anybody can use it via a self-service model and access the AI database. The advanced ML capabilities can identify suspicious access as well.
“We use the same infrastructure for managing autonomous database. A lot of the mundane tasks go away through ML. We look at the matrix, logs, events happening in the database and we can detect anomalous patterns. We can find out if something does not look right and detect anomalies in advance using those techniques,” he contends.
According to Lahiri, if a problem happens, they can find out if it matches an existing problem or if there is an existing bug that fits the problem description and therefore can be applied automatically. There can also be a new problem that doesn’t match an existing symptom, so the system will immediately file a new service request and escalate it to get a proper response and apply a security patch automatically.
“We also use Artificial Intelligence (AI) techniques to optimize the Autonomous Database. As the database runs, we use AI techniques to decide, whether we have the right access structures in the database, should we change the indexes that we have in the database, because the workload now has a different profile from yesterday,” wraps up Lahiri.