Posted May 25, 2016 By Susan Hall
Database-as-a-service offers a way for companies to better handle data growth and manage multiple databases.
Thanks to huge growth in data volume, as well as a proliferation of new data sources and data types, enterprises are scrambling to meet pressing data storage and processing needs. Database-as-a-service (DBaaS), which puts data storage and management in the cloud, is being hailed by some as a great way to solve these vexing data management issues.
Not the Same Old Database
Rather than having a single database technology, enterprises increasingly maintain a “polyglot persistence,” in effect employing multiple data storage technologies that meet the needs of various applications.
Today’s applications can involve multiple databases. A smartphone app to help drivers find a parking spot, for instance, might involve an inventory, a mapping tool, a payment process and a customer relationship management tool. Today’s databases aren’t just document stores, either. They might include graphs, geospatial data and more. And each of these databases might have unique maintenance requirements.
DBaaS offers a way to simplify the management of these multiple databases. At the same time, enterprise IT departments are procuring more services through the “anything-as-a-service,” or XaaS, model, focusing their talent on innovation and service delivery instead of nuts-and-bolts technology infrastructure.
Among the factors fueling the DBaaS trend, according to Frost & Sullivan:
- Ability to better manage data growth. DBaaS enables provisioning of new databases quickly, securely and cheaply.
- More flexibility/agility. The ability to quickly scale up or down to meet needs of data flow can be more cost effective than buying hardware to meet peak demands – especially when the boxes remain idle at other times.
- Improved resiliency
- Simplified management. Rather than managing a complex collection of database silos, the organization can employ a data tier that is distributed and diverse while leaving data provisioning, configuration, patching, tuning, monitoring and security to a vendor partner. By automating these tasks, DBaaS can provide centralized management for a complex web of systems.
The DBaaS market is expected to be worth $1.8 billion in the coming year, according to a 451 Research report by analyst Matt Aslett.
Companies collect data from a variety of sources, including connected devices, employing it for BigData analytics, security and compliance initiatives, among other uses. They’re looking for solutions that are flexible, can scale, provide high availability and can process queried information at lightning-fast speeds, according to Frost & Sullivan.
However, database-as-a-service providers differ in their capabilities, so companies interested in DBaaS need to carefully consider their goals, requirements, fault tolerance and expected costs, according to Frost & Sullivan.
Picking a DBaaS Provider
Incumbent database giants Oracle, Microsoft, IBM have been developing DBaaS solutions, and are expected to hold their own against a wave of DBaaS startups, according to 451 Research.
Respondents to a 451 Research survey had, on average, 2.9 database providers. They most commonly used a private DBaaS for custom production applications, while they cited online transactional processing (OLTP) among their top uses of public DBaaS. They used both for application development, web applications, online analytical processing (OLAP) and line-of-business applications such as CRM.
A public DBaaS service might even involve multiple vendors, points out 451 Research, noting that the AWS Relational Database Service supplies Oracle Database and Microsoft SQL Server instances, as well as MySQL and PostgreSQL.
“DBaaS services tend to be ‘sticky,’ due to the difficulty of migrating large amounts of data to or from the cloud. As such, it is an attractive service for cloud providers that seek to move up the cloud stack with value-added cloud services,” states the Frost & Sullivan report.
Many technology providers also heavily market DBaaS offerings through value-added resellers (VARs), managed service providers (MSPs) and system integrators (SIs).
Companies are most likely to choose a DBaaS vendor based on its ability to integrate with existing production applications, support for specific databases, security, and ease of use for end users, according to Aslett’s research.
Of course, it’s also important to consider the price. DBaaS vendors have begun offering more choices, including per-minute pricing models. Amazon’s Aurora product, for instance, offers granular pricing depending on how much storage you want to buy in advance, how much you want to buy on-demand during testing and how much backup you want to store. International costs, such as Japan’s consumption tax, could apply, too.
Possible DBaaS Drawbacks
The biggest DBaaS drawbacks Aslett sites are security concerns, migration/integration concerns, interoperability issues and lack of clarity on cost. He noted a drive toward installing a fully-cloud-based application suite and company mandates to use cloud services to cut costs and maintenance for existing installed databases.
Acceptable levels of latency can be another issue. Traditionally, the database and the application server were in the same location, reducing network latency for database requests and allowing both to be protected by the same firewall. Not so for public distributed systems.
The location of data was among DBaaS concerns in research from database security firm HexaTier. Cloud providers maintain data centers in multiple locations to provide redundancy and improve performance. But some governments, especially within the EU, require data be kept within their physical borders. And the physical location of data can be an issue with some compliance mandates.
The 574 IT pros it surveyed also were worried that multi-tenancy might allow outsiders to access the data, that the vendor would fall victim to a cyber attack or that a vendor’s authorized users might pose a threat.