Aiven powers OVHcloud’s growing DBaaS portfolio, Managed Cloud Databases. Read this blog post from Aiven CTO, Heikki Nousiainen, to find out how their customers have achieved better agility and efficiency in the cloud – and how you can do the same!
Why business agility is important
For companies in today’s changing world, it’s crucial to remain agile and adapt to changing circumstances. Failure to change directions, sometimes in mid-step, can result in needless expense and hurt a company’s image.
When we talk about agility, we often think of product and corporate strategy and other long-term progressive endeavors. That’s valid, since both the product roadmap and the overall strategy of the company must be able to change to fit the market environment, as well as in response to internal developments.
But agility is also relevant in operations and infrastructure, where a company’s internal workings must also be able to accommodate change. New innovations come to market every day that can save money in your operating expenses; and every now and then tools become obsolete, even though you’re still using them. If a company doesn’t react in a timely manner, these changes can result in reduced efficiency or actual loss of business.
What’s more, any company’s operations and infrastructure absolutely must support innovation on a fast cycle, or compromise their ability to adjust to change.
DBaaS in the cloud makes sense
One of the pieces of infrastructure that it most makes sense to purchase as a service is databases and related services. Setting up and managing a database is a specialized skill, especially doing it well, and requires resources that can instead go into developing a company’s own products. In practice that means that instead of hiring a database admin, you can hire a developer for your product team, while another company with its own skilled team is taking care of your data, and you can transfer that slice of budget from OpEx to development investments.
On a more general level, data is one of a company’s most valuable assets, and plays a key role in all modern online and software products. This means not just data as code and images, and not even the data that flows when the service or product is being used. Storing and analyzing customer and usage data can tell companies how their business is doing and how they should be directing future development efforts. Building the capacity for this data storage is an investment that’s hard to scope out in advance, and analysis needs will evolve as the business grows and the market changes.
Open source for freedom
Things change all the time, and whatever service you buy may not be around for as long as your company, or they may change their terms without consulting you. Or you might just outgrow their offering and need a new provider.
When buying your data infrastructure as a service, the one thing you want to ensure is that you’ll always have access to your data, even when you decide to switch services. The data also needs to be in a format that’s compatible with other providers and products, or you stand to foot a massive bill for conversion (if it’s even possible).
Here’s where open source is key. If you go with a service that offers fully open source products, your data will always be in a format that’s easy to export elsewhere, and it’ll be compatible with a wide variety of solutions. Open source gives you real freedom and flexibility to choose your providers and not be locked into a solution offered by a single company.
Shortening the innovation-development-testing cycle
Managed cloud databases are increasingly the key factor that enables quick innovation. Product development teams need a way to quickly turn ideas into testable products, and ready-made, managed infrastructure is the only way to achieve this in the IT world. The more building blocks you can buy ready-made, the quicker you can sketch out a test version of your dreams, and the quicker the roll-out when it’s finished.
DBaaS customers can provide data-driven services without having to build up the expertise in managing a new system and then having to build a datacenter. Their developers can quickly get to the part where they start programming, which greatly shortens the time to market. Being ahead of the curve by months can make or break a product, no matter how high its quality otherwise.
And speaking of quality–having the means to spin up data infrastructure quickly also means a company can continuously improve their products. With a DBaaS-sourced architecture, new environments can easily be set up for testing.
This is equally true when the issue is not just improving upon or enlarging existing services, but creating entirely new products. The road from idea to product is much shorter when the infrastructure is already built.
And lastly, it also means many teams can work on the same codebase simultaneously. This way the company can make every R&D penny count.
From CapEx to reduced OpEx
Then there’s the fact that datacenters cost money to build, and you’re using your CapEx to do it: it’s an investment in the most classical sense. Even if you’re leasing server space instead of building walls, you’re still spending your OpEx coins in a predefined amount of capacity and its maintenance.
The amount of capacity a company maintains has to accommodate not just everyday use but also traffic spikes. When an investment is made (usually in some hurry) to increase capacity in response to an emergency need, it’s seldom considered safe to let go of it when the emergency passes because the process is painful and slow. This road usually ends with services overprovisioned by some 30-50% for regular use. And this is true whether you’re doing it with CapEx or OpEx.
CapEx reduction is easy to achieve by moving to a managed service, and there are plenty of them. But by selecting the right DBaaS, a company can make sure that their capacity meets their needs on a day-to-day basis: any service worth its salt will make it easy not just to scale up but also to scale down. It’s possible to reach a 70% reduction in infrastructure OpEx by managing your capacity and infrastructure properly. At the same time it lowers the threshold for using these services for new products, as well as retiring products that aren’t generating significant revenue.
Orchestration as a multiplier
There’s an African proverb: If you want to go fast, go alone; but if you want to go far, go together. Creating a standalone product to serve a single purpose is quick, but to provide real long-time value for your customers and your own company, versatile products or services that offer a range of functionalities and meet multiple needs are key. You might hit a goldmine by creating a single-purpose product or service that satisfies an urgent user need that no one else in the world has thought to satisfy, or you do it better than absolutely everyone else. It’s just not very likely, to say the least.
A more robust business model is to target a wider customer base and meet a variety of needs. Versatile products are more useful to a larger group of people, and customers are more likely to want a solution that meets several needs at once. Even if you start small with a tightly targeted product or service, you want to keep the door open to expansion and innovation available in adjoining spaces.
The reason that products like this are harder to create and hence more valuable is that it means creating a heterogeneous data architecture that encompasses a wide range of building blocks. This is not easy, whether you’re building them in-house or buying them as a service.
And making the components in your infrastructure all work together and pass data along flexibly is actually really hard. Companies that buy their data infrastructure as a service may have it even harder: using several services from multiple providers can add to the complexity of the overall architecture, and the range of available services can limit what it’s possible to accomplish.
Note that the data flow itself is not an issue, since that can be configured; in some cases easily, in some cases less so, but it can be done, and then it’s only a question of how easy it is to add data sources.
The bigger issue is the regular management and maintenance tasks, which can be a pain if you have to log into and out of multiple environments to check service status or perform updates. If your system management is fragmented, it’s hard to acquire an overall picture of the whole system, and that has implications for troubleshooting and optimization.
Companies using orchestration systems like Kubernetes and Terraform to centralize and automate management tasks find that it enhances their control over resource consumption. With an orchestration system, companies can utilize a mixed set of services as easily as just one. These systems make it possible to centrally manage both individual services and the way the services interact. This gives a better overview of system status and performance and makes it quicker and easier to troubleshoot and fix issues. And of course, access management can also be centralized to an orchestration system, removing yet more regular upkeep tasks.
Kubernetes and Terraform also provide greater horizontal scalability. Instead of having to build endpoints into each service individually, it’s enough to build them in the orchestration system–and both of them also come with a wide selection of ready-made open source connectors, saving companies the trouble of writing their own.
This is how the benefit from orchestration combines with the benefit derived from increased agility in the marketplace and takes both to a whole new level. At Aiven, we see this all the time. Large numbers of our customers use Kubernetes and/or Terraform to save time, money and effort in terms of both system maintenance and product development.
The common denominator for companies adopting managed cloud databases and combining it with an orchestration system is that they’re able to reduce their expenses markedly, due to a combination of reduced capital expenditures and smaller operating costs, even when taking into account the fees paid for their database management tasks.
With a DBaaS in use, companies can also stay agile and responsive, and continuously evaluate and adjust their business model as the circumstances change. With OVHcloud’s managed cloud databases, powered by Aiven, you don’t have to worry about your database infra or administration, including set up, backups, scalability and security. This saves you time and money, so you can focus on building and deploying the cloud-based applications you want.
Heikki Nousiainen is Chief Technology Officer and one of the founders of Aiven, a next-generation managed cloud services company offering the best Open Source data technologies to empower businesses around the world and make developers’ lives better. Having raised $210M at a $2B valuation and scaled Aiven on three continents with headquarters in Helsinki and hubs in Berlin, Boston, Paris, Toronto, Sydney and Singapore, Nousiainen has a vast knowledge of technology, open source and licenses, IoT and more. He previously worked as a software architect, information security specialist and software engineer for computer software and IT companies, especially driving cloud adoption.