NuoDB: The Elastic SQL Database
This video demonstration shows how easy it is to elastically scale your NuoDB database, remain resilient to failures & outages, and deploy in an active-active hybrid cloud environment.
Welcome to the demonstration of NuoDB - the Elastic SQL database for always-available, always active, hybrid cloud applications. In this demonstration you will see how easy it is to elastically scale-out your applications by simply adding a node.
When you simply add an in-memory transaction processing engine, you will see throughput - as measured in transactions per second increase, while latency drops. We can keep adding transaction processing nodes to dramatically increase transactions per second - in this case all the way up to 2.8 million transactions per second!
In addition, we will show you how you can deploy on-premises, in the cloud, or across a hybrid cloud. And when deployed in an active-active configuration, if your primary data center suffers an outage, your application continues running with no disruption to application service.
In this extended demo, you will see how NuoDB brings you the benefits of cloud deployments and elastic scalability, without sacrificing the requirements of business critical, operational workloads.
Easy, Elastic Scaling
Let’s start by having a look at how easy it is to elastically scale-out with NuoDB
In this demonstration NuoDB is running in Red Hat OpenShift, and throughput and latency are being visualized in simple graphs. In the lower left we see a logical topology of the NuoDB deployment.
In the simplest NuoDB configuration we have one in-memory transaction processing node - called a Transaction Engine - and one storage manager for durability. If we want to increase resiliency, we can simply adding a 2nd storage manager node. NuoDB then automatically syncs storage data from other nodes and we get full storage redundancy.
As the application workload increases though, we notice that client latency begins to increase as well. So we a simply add a 2nd transaction engine, increasing throughput and reducing latency.
It’s that easy to scale-out with NuoDB!
Database Resilience & High Availability
Now let’s explore high availability with NuoDB.
NuoDB is resilient to failure of any node. In this case we have 6 nodes running, including 4 transaction processing engines. Unlike traditional monolithic databases, if a transaction engine goes down, NuoDB automatically reconnects the application to another transaction engine. While throughput may decrease, the application keeps running and we can easily scale up throughput again by simply adding back another transaction engine.
Active-active | Hybrid Cloud
Another unique aspect of the NuoDB architecture is its ability to be deployed active-active across a hybrid cloud architecture.
Let’s start by running NuoDB in a typical on-premises deployment. If we want to expand to the cloud, we can simply add a second storage manager in our cloud environment giving us additional storage redundancy and resiliency.
We can then add transaction engines to our cloud deployment, spreading the workload across the on-premises and cloud environments, and as we connect our cloud application, we end up with a fully hybrid deployment spanning on-premises and public cloud.
We can continue to simply add nodes to our cloud deployment to increase resiliency, capacity and throughput. And if we see our cloud workload increase, watch how OpenShift and NuoDB automatically add additional transaction engines to accommodate this workload, increase throughput and reduce latency.
Continuous Availability | CLoud Migration
Now let’s look at how NuoDB supports continuous availability and application migration to the cloud.
With our hybrid deployment spanning on-prem and the public cloud, if we shut down our on-prem data center - or it suffers a catastrophic failure, the application continues to run in cloud uninterrupted. Throughput may drop, but the database - and the application - keeps running, allowing you accommodate data center failures and migrate applications to the cloud without an interruption to application service.
Cloud scale, on-demand
Finally, let’s look at how far you can scale-out your application, on-demand.
As we’ve already seen, with NuoDB you can easily scale-out and increase throughput by simply adding a node. As we add more in-memory transaction processing nodes, watch as our throughput exceeds 2.8m transactions per second, easily exceeding the throughput requirements of most applications!
In this demonstration, you’ve seen how NuoDB uniquely provides cloud deployment benefits while maintaining relational, transactional guarantees.
If you’d like to experience an elastic SQL database for yourself, download the NuoDB Community Edition at nuodb.com/download.