Businesses looking to innovate and transform cannot risk database downtime. High availability for your most mission-critical applications, workflows, and clusters is one of the most important success factors for any organization looking to become a leader in their market space.
But what does true high availability look like? How do you achieve it? What database solutions can ensure it—not just when you need it, but all the time?
In order to answer these questions, EDB turned to our team experts: Kevin Li and Michael Willer. Kevin Li serves as EDB’s Director of Sales Engineering for EMEA and Michael Willer as EDB’s EMEA Senior Sales Engineer. Together, they sat down for an Ask the Expert webinar on June 9, entitled “Extreme High Availability with PostgreSQL.”
Over the course of an hour, Li and Willer dove into topics such as:
- Why does high availability matter?
- What solutions are best for high availability?
- How do EDB solutions deliver high availability?
- What architectures are best suited to high availability?
For those who were unable to attend this webinar, we wanted to compile some of the most valuable takeaways. With these insights and the full recording, you’ll be in a better position to ensure that your database is truly ‘Always On.’
Defining high availability
When we talk about high availability for a database, whether on-prem or in the cloud, the defining factor is preventing unnecessary downtime while optimizing uptime.
Every database will experience scheduled or planned downtime once in a while—whether to install updates or fix bugs. However, these processes are frequently allotted to times when an organization is confident traffic on their site or use of their applications is low. In short, while scheduled downtime is inevitable, it needn’t be disruptive.
Unscheduled (or unplanned) downtime, however, is a major issue for enterprises. Internally, it disrupts your teams’ ability to get their work done, to access the tools and data that they need. It leaves IT scrambling to figure out where, what and why the central infrastructure of your organization is down.
This is where a highly available database becomes most valuable. High availability ensures that your organization can minimize (if not erase) the unplanned database downtime that might stymie your ability to accomplish projects or serve customers, users or partners.
Why does high availability matter?
The cost of unplanned downtime can be severe, both internally and externally.
Internally, your database is the beating heart of your business. It provides all the necessary resources and information to your most vital teams. Should your database regularly fail, your ability to achieve both day-to-day tasks and large-scale projects will fail as well. And you’ll see the results externally.
Excessive downtime diminishes your users’ and customers’ trust in your brand. It leaves them feeling underserved and overstressed. Not only does downtime make it harder to attract new customers, but—more severely—it drives existing customers away, no matter how loyal they might have once been.
This has an immediate impact on revenue—and a far larger one than many enterprises realize. By EDB’s calculations, businesses can lose up to one million dollars of revenue per hour of downtime, per year (even higher for banks and trading platforms). This means that, even if your database is available 99% of the year, down for roughly 87.6 hours annually, your organization could be missing out on an additional $87,600,000 of revenue per year. However, if you invest in a database that offers the “Five Nines” (99.999% uptime) of high availability—less than 6 minutes of downtime per year—that number drops to just $87,600.
For some organizations, the risk of lacking a highly available database can be even more dire. Healthcare organizations, police departments, national security agencies, hospitals and fire brigades have more than revenue on the line. In the case of organizations such as these, ensuring constant access and optimal uptime for a database infrastructure can, without exaggeration, become a matter of life and death.
How does EDB and Postgres provide high availability?
As the most loved open source database among developers, Postgres is renowned for its high availability. A large part of this stems from the global Postgres community, which constantly works to maintain, support and evolve the Postgres database project. In short, this means that Postgres has tens of thousands of experts working in every time zone to make sure that the database is running.
Not a bad support system!
At EDB, we’ve worked to build on that strong foundation, with what we call “extreme high availability.” This capability is provided by EDB Postgres Distributed, a PostgreSQL extension providing multi-master replication and data distribution with advanced conflict management, data-loss protection, throughput up to 5X faster than native logical replication, and enables distributed PostgreSQL clusters with high availability up to five 9s. EDB Postgres Distributed is compatible with Postgres, EDB Postgres Extended Server, and EDB Postgres Advanced Server distributions, as well as BigAnimal, our fully managed Postgres database-as-a-service. Our dynamic solutions provide advanced failover mechanisms that ensure your database is up and running to the Five Nines standard. Should your database go down, EDB failover capabilities can immediately detect it and switch the database over to another node, so it’s constantly available. Additionally, EDB provides constant support, backed by members of Postgres’ own Core Team.
EDB’s extreme high availability provides:
- Five Nines of database high availability
- Fast automated failover
- Minimal data loss
EDB and Postgres: an always-on database
Your database needs to keep up with your business, your ambitions and your users; otherwise, it’s no good for you. That’s why high availability must be front of mind for any organization looking to innovate and transform.
While we couldn’t share every piece of insight from this Ask the Expert session, we hope you’ll watch the webinar in full to learn not only why high availability is mission-critical for your enterprise, but how you can achieve it in a way that suits all of your organizational needs.