ScaleArc

Active/active Data Centers

Enable continuous availability. Span prem and cloud.

Ensuring continuous availability for critical applications has become a top business priority. Companies want active/active operations, but the cost of this architecture has been too high to justify, given the sophisticated application, database, and networking configurations needed. Companies have instead lived with active/stand-by architectures.

ScaleArc makes it much easier – and far more affordable – to architect for active/active operations. The ScaleArc software provides a range of traffic management capabilities, with four key features that make it possible for organizations to easily adopt and benefit from active/active operations:



Automatic support for read/write split
ScaleArc has the ability to understand which queries are reads vs. writes and then appropriately distribute the traffic accordingly, sending all write transactions to the primary server and load balancing all reads across the available secondary servers. With ScaleArc, you don’t need to modify your applications with configurations such as read-intent strings to get the performance and reliability benefits of read/write split.

Geo-aware load balancing
ScaleArc uses Time To First Byte to understand server response times, so the software will send queries to the servers that are closer and will send more load to those better-performing servers automatically.

Replication-aware serving of data
ScaleArc monitors replication lag through a replication monitoring engine. The software will automatically avoid sending traffic to a node that has fallen behind the user-set threshold for replication delay.

Auto failover within or between data centers
ScaleArc leverages the failover architected in the database to provide automatic failover at the application layer. The software recognizes a failover is underway, queues inbound queries to avoid application errors, and then forward the queries once the secondary is promoted to primary. The software can also route read queries to other available secondaries throughout the failover, ensuring continuous application availability. Because the application is directed to communicate with the ScaleArc abstraction layer, changes in the database structure behind that layer are transparent to the application, enabling cross-data center failover with no application changes.