Database As a Service. Q&A with Tim Rottach
Q1. What are the main benefits of a Database As a Service (DBaaS)?
One of the biggest benefits of a Database-as-a-Service for customers is the shifting of responsibilities of database management to the service provider and the streamlining that occurs through automation. This means less operational effort and cost for customers and automation also prevents human error. A fully managed service makes it easy for customers to set up a new database, control several databases from one control plane, evolve database services and associated hardware over time, and automatically update their databases. Additionally, customers leverage the power of the cloud to help improve high availability on a global basis. Also, an important benefit related to automation is improving security, as patches and upgrades can happen centrally to protect all customers at the same time.
Q2. What are the main challenges when implementing a DBaaS in an enterprise?
Like most types of cloud-based software, cost can be a big factor. That is why price-performance is a key factor in choosing a DBaaS. As customers scale applications, the demand on the database grows and high-performing and scalable architecture becomes critical to keeping costs in line. Additionally, multi-model DBaaS solutions can deliver for customers across multiple use cases vs. individual point solutions. This means customers can reduce their overall number of service providers, reducing the total cost of ownership. Flexibility is another challenge as applications evolve quickly with modern agile development. A DBaaS must support a flexible data model to change with the application’s needs. Security and data governance are always a concern with any database, but more so for companies when the service provider has a larger role to play in this area.
Q3. There is a claim that “Cloud databases enable organizations across every sector to perform the functions that traditional databases do but, at an accelerated pace and with more flexibility and scalability”. Can you please comment on this?
Cloud databases have more automation built into them by their nature. They provide more convenience and allow customers to “spin up” and “wind down” dev and test environments quickly. This provides great flexibility. Evolving database clusters, services used, cloud regions, and data movement can all happen faster in the cloud because the infrastructure is already in place from cloud providers and doesn’t need to be purchased, set up, and configured by customers in advance.
Q4. What specific challenges pose connected IoT devices?
With IoT scenarios, applications can easily use tens if not hundreds of thousands of devices. Each device can be sending a large and constant flow of data back to a centralized cloud database. Devices for an application can of course have a variety of types that may change how the data is captured. A database platform that can overcome these challenges must have an embeddable cross-platform database that is the same as the centralized cloud database and have sophisticated data synchronization technology to streamline and manage the flow of data.
Q5. Please tell us in a nutshell what is a NoSQL cloud database?
A NoSQL cloud database is a non-relational database that works in the cloud. NoSQL databases are well suited for handling large data volumes with high performance and providing flexible data models that allow teams to quickly evolve application design as their needs change. Rigid data models are a common limitation of relational database design. Though not relational in nature, advanced NoSQL databases have similar capabilities to relational systems, including using SQL as a query language, cost-based optimization, user-defined functions, and supporting distributed ACID transactions to name a few.
Q6. What specific features are useful to handle in the future zettabytes (ZB) of data?
With each year we are producing and using more and more data at an accelerated pace. To manage huge volumes of data it is critical for organizations to understand what parts of this data is the most valuable and which data will play a lesser role. The most important data should be processed quickly in-memory on the best hardware while less important data can move more slowly and be stored for a more limited time. Organizations will not want to keep all of their data forever. It is about what is the most important data and how they will use it. Flexible data systems will support sophisticated data management for different tiers of data.
Q7. Anything else you wish to add?
While cloud databases are a critical and rapidly growing area, some enterprises still retain control over their most important systems. For Enterprise-grade database providers, it is essential to have both a multicloud DBaaS platform, support customers’ self-managed solutions, and give customers a choice of hybrid cloud. Enterprise customers seek solutions with great scalability, flexibility, price performance, and security.
…………………………………
Tim Rottach, Couchbase director of product marketing.