No results found

Your search did not match any results.

Cloud | Database Innovation

Oracle Database Is Designed for the Cloud

Oracle Database Is Designed for the Cloud

By John Soat

 

Here's what makes Oracle Database as a Service such a compelling option.

Database technology has always been one of the critical underpinnings of enterprise IT, and for that reason databases have typically run in the corporate data center. But new capabilities in Oracle Database 12c now make it feasible for organizations to run their databases in the cloud, providing much greater flexibility to businesses and freeing up IT resources for use in customer-focused innovation.

Larry Ellison expects cloud-based database services to grow quickly in popularity. “Database is our largest software business and database will be our largest cloud business,” Oracle’s Executive Chairman and Chief Technology Officer said in September at Oracle OpenWorld 2014.

Oracle Database 12c is based on a multitenant architecture that lets businesses quickly and easily implement “pluggable databases” that exist within a container database. This innovative design makes it possible for IT teams to offer database as a service (DBaaS) with unprecedented flexibility.

There are three main layers of cloud computing: software as a service (SaaS) refers to applications in the cloud; platform as a service (PaaS) includes databases, middleware, and programming tools; and infrastructure as a service (IaaS) provides compute power, storage, and networking.

Main Layers of Cloud Computing

Saas, PaaS, and Iaas are the three main layers of cloud computing.

Oracle offers a complete set of services at all three layers of the cloud, and has been providing database cloud services for a few years. Its original offering, Database Schema Service, allows a user to access a portion of an Oracle Database running on Oracle infrastructure. “You get a dedicated schema, a slice of the database,” says Dom Lindars, Oracle Cloud senior director of product management.

Addressing Database Sprawl

What’s different now, and what Larry Ellison introduced at Oracle OpenWorld 2014, is the use of a complete database instance in the Oracle Cloud, with a subscription service model. Customers can access a fully functional Oracle Database running on a dedicated virtual machine (VM), with full administrative control, quickly and efficiently. It exemplifies the benefits most closely associated with cloud computing: agility, scalability, and cost effectiveness.

It also addresses a major area of concern to IT professionals: database sprawl. “We have many, many customers that have more than 10,000 databases,” says Lindars.

Large enterprises, and even many midsize ones, can have thousands of databases, the majority of which are created to help developers build and test applications before moving them into a production environment. Those multitudinous databases represent significant overhead in terms of management, resources (both hardware and personnel), and security.

Database Sprawl

DBaaS can help contain database sprawl.

DBaaS, in tandem with server virtualization, helps address sprawl by corralling databases within a consolidated cloud infrastructure, thereby reducing database hardware overhead and improving data security. That helps explain the burgeoning interest in DBaaS, the market for which is estimated to grow from $1 billion in 2014 to $14 billion by 2019, according to MarketsandMarkets, a research firm.

Oracle Database 12c’s multitenant capability, Oracle Multitenant, turns out to be “a very efficient architecture for deployment,” says Patrick Wheeler, senior director, Oracle Database product management. “It’s about the granularity of the equipment you can provision in the cloud,” he adds.

Oracle Multitenant lets DBAs consolidate database instances using container databases. Up to 252 pluggable databases can run in a single container database—one container database per VM, says Wheeler. That translates to dramatic hardware savings over and above server consolidation.

Background Efficiency

The container approach is more efficient than non-multitenant databases running on VMs, Wheeler points out. Every database system performs myriad background tasks that keep the database up and running. Multiply those tasks by the number of database instances running on a single core in a virtual environment and it makes for a performance bottleneck. “Database loads don’t do well when they’re multiplied,” Wheeler says. The container model, which represents a single set of background tasks for up to 252 databases, “is much more efficient,” he says.

In the same way, Oracle Multitenant enhances security. Each pluggable database represents a fully isolated application dataset. Also, a pluggable database inherits the security incorporated in its container database, so patching is easier and more effective. And, because a pluggable DB can be unplugged from its container and re-plugged into another, upgrading security is as simple as unplugging one and re-plugging it into a security-enhanced container.

Oracle Multitenant makes it possible to create new databases very quickly, including cloning databases that already exist, a critical function for developers. Experienced users can get a database “up and running in an hour,” says Lindars, and employ that database “for an hour or two, or for months at a time, or for years and years.”

DBaaS Management

Customers who use Oracle DBaaS have access to Oracle’s extensive database toolset, including Oracle Enterprise Manager 12c, which automates control of the pluggable databases in a container database. Not only does Oracle Enterprise Manager 12c test and tune the databases, it provides metering and chargeback capabilities that let managers monitor and report database usage with granularity, an increasingly important function in service-oriented private cloud architectures.

Two other tools enhance the uptime and recoverability of Oracle DBaaS. Oracle Real Application Clusters (RAC) creates a scalable database environment across clusters of servers and passes workloads to available processors. Oracle Active Data Guard maintains backup copies of databases, including, with Oracle Multitenant, pluggable databases, that can be tapped in case of system failure. Oracle Engineered Systems provide a significant competitive advantage in the DBaaS model. Oracle Exadata Database Machine and Oracle SuperCluster are optimized for database workloads. That makes these systems particularly well- suited for high-performance private clouds.

Finally, Oracle has invested considerable effort to ensure the easy portability of its technology, which serves customers well in the DBaaS model. With Oracle’s DBaaS, developers can build, test, and tune applications using easily created or ported or cloned pluggable databases, then migrate that workload to an on-premise system running an Oracle database in production. “If you build things on top of the Oracle Database, you can run them anywhere you want to run them,” Lindars says.  


Get Started