Experience MySQL scalability and high-availability


Benefit MySQL NoSQL Oracle RAC ScaleDB
ACID Compliant Yes Yes Yes
MySQL Compatible (Tools too) Yes Yes
Compatible with Public Clouds Yes Yes Yes
Scale-Out without Sharding Yes Yes Yes
Cluster-Level Load Balancing Yes Yes
Highly-Available Yes Yes Yes
Dynamic Elasticity Yes Yes Yes
Database Virtualization Yes
Set-up a Cluster in an hour Yes
Eliminates the Need for Slaves Yes Yes Yes
Move Cloud Instances On-The-Fly Yes
Scale Instance Size On-The-Fly Yes
Scale I/O and Compute Separately Yes Yes


ACID Compliant: ACID is an acronym for Atomicity, Consistency, Isolation and Durability. Read more here. NoSQL databases use a relaxed (or eventual) consistency model, so they are not ACID compliant.[back to top]

MySQL Compatible (Tools too): MySQL has the largest installed base of all databases. It also enjoys a huge ecosystem of tools and applications. For these reasons, MySQL compatibility, at the application and tool level, is often considered important. ScaleDB actually uses and extends MySQL, so obviously it is not only MySQL compatible, it is MySQL.[back to top]

Compatible with Public Clouds: Flexibility to deploy your application on any Linux-based cloud (public, private or hybrid) or on bare metal machines is important to keep your deployment options completely open.[back to top]

Scale-Out without Sharding:Sharding your database involves creating a collection of independent databases. This requires that your application route each database call to the appropriatedatabase. It also requires that your application handle all functions involving more than one database shard. Sharding can be a painful process, but resharding is far more painful. Scaling-out MySQL, without ScaleDB, requires sharding.[back to top]

Cluster-Level Load Balancing:When you partition or shard your database, each database contains a unique piece of data and must handle the load for that data. Shared-data databases like Oracle RAC and ScaleDB enable all nodes to process the entire collection of data. This is best described with an example. If your database is very busy with payroll on the 15th, logins every morning and expense reports on the 30th, then a sharded database would require that each server handle its own peak load for these things. On the other hand, a shared-data database running four-nodes could have all four nodes running payroll on the 15th, logins every morning and expense reports on the 30th, because each node can handle any database request.[back to top]

Highly-Available:MySQL, set-up in a master-slave configuration provides fail-over, whereby a slave is promoted to the master role. This process takes time and typically involves application downtime. High-availability requires that the application keep running.[back to top]

Dynamic Elasticity:The ability to scale both up and down, by adding or removing nodes without interrupting your database operation is dynamic elasticity.[back to top]

Set-up a Cluster in an Hour:ScaleDB and Oracle RAC both operate as a cluster, a collection of machines addressed as if they are a single machine. You can be up and running a ScaleDB cluster in less than 30 minutes, while Oracle RAC typically takes days or weeks. [back to top]

Eliminates the Need for Slaves:MySQL, in order to provide fail-over, relies on replication to slaves. ScaleDB and Oracle RAC do not employ slaves since they provide a master-only clustered high-availability that is more robust than master-slave replication. [back to top]

Move Cloud Instances On-The-Fly:This is particularly applicable in cloud environments, where there is a sharing of resources (compute, RAM, storage, network, etc.). Quality of service (QoS) can suffer from noisy neighbors, whether in the same physical server or on the same network segment. ScaleDB enables you to add an instance, which is on another server or network segment, to the cluster and then remove the one that is suffering from the noisy neighbor. In effect, you are adding a new node to the cluster and then removing the old troubled node.[back to top]

Scale Instance Size On-The-Fly:When running in the cloud, you often need to move to a larger or smaller instance size. In a public cloud, you don‘t even know whether the new instance is on the same server or a separate one. ScaleDB enables you to add the new instance, whether it is larger or smaller, to the cluster and then simply remove the old instance. This is how you can change the instance size of your database on-the-fly without downtime.[back to top]

Scale I/O and Compute Separately:ScaleDBseparates and virtualizes the database compute and the database storage functions. As a result, if your database is CPU-bound, you can add more compute nodes. If your database is I/O bound, you can add more storage nodes. This is one of the core benefits of true database virtualization. [back to top]