Grow with confidence

Scale-up vs. scale-out — What makes an optimal scaling solution?

Tony Branson
© Shutterstock / pathdoc

Scale-up or scale-out should not be a dilemma for any enterprise – the choice is very simple. In this article, Tony Branson explains how organizations can grow their infrastructure with superior performance, high availability, and agility.

Superior performance, high availability, and scalability make a great foundation for an agile infrastructure that is fully equipped to handle the increasing complexity of operations while keeping up with the ever-evolving user needs. Building such a system is no trivial task, so it is important that you understand why performance, scalability, and availability are vital to implementing robust and agile business solutions.


Scalability refers to the capability of an application or a website to expand in response to increasing workloads and growing demand for processors, network or file system resources. Any system can be scaled horizontally by adding new nodes that have similar features for uniform redistribution of load and vertically by adding storage, memory, or processors to host extra virtual servers using the same hardware.


System performance means the throughput recorded in a specific timeframe for a given amount of workload. Performance is typically validated by testing the resilience and scalability of network, software, and hardware.

High availability

High availability describes the ability of a system to ensure that applications hosted in the middle tier remain up and accessible for the clients around the clock. It can be achieved through the redundancy of multiple database servers, where the cluster continues to operate even while one or more servers face downtime or become disengaged from the network.

SEE MORE: Elements of Scale: Composing and Scaling Data Platforms

HA requires the database to support automatic failover from one primary server to another. However, to ensure application availability, the database failover is best complemented by database load balancing software. This software will queue writes during failover and drain the queue when the failover is complete. So users see delay but not errors.

Scale-up vs. scale-out: What makes an optimal scaling solution?

Over optimization is a commonly made mistake that invites unnecessary cost and complications so scale up or scale out should not be a dilemma for any enterprise – the choice is very simple. Scaled-out architectures enable the system to grow with growing business needs while leveraging lower-cost components. Horizontal scaling efficiently accommodates a large number of nodes in a single management entity to ensure that performance does not decline.

The challenge is having applications be able to take advantage of horizontal scale out at the data tier. Database load balancing software enables scale out without code refactoring or making any changes at the app level. It is designed to facilitate instant scale-out that also increases application availability, efficiently meeting the growing needs of the end-user.

Implementing scalable systems with a load balancer for better throughput and fault tolerance

High availability, scalability, and performance can be incorporated into every layer of the technology stack – the challenge lies in having applications tap into those capabilities at the data tier. Implementing database load balancing software lets you architect your infrastructure for improved resilience and business continuity performance. It distributes queries across multiple servers and instantly routes traffic, shielding apps from database downtime during seasonal traffic spikes.

SEE MORE: What you need to know about enterprise-scale continuous delivery

With database load balancing software, DBAs can scale up in an instant and scale out transparently. It ensures complete data recovery following a disaster, keeps mission-critical apps up and running even with high traffic volume, and facilitates automatic failover without disrupting the end-user experience. Enterprises that do not want to spend their time, effort and resources on building expensive, unnecessary efficiencies can easily achieve high-availability, database scalability, and performance with database load balancing software.



Tony Branson

A self-proclaimed tech geek, with a passion for ScaleArc’s disruptive technology innovation in database load balancing. Tony has a passion for dissecting tech topics such as transparent failover, centralized control, ACID compliance, database scalability and downtime effects. On his days off, he can be found watching sci-fi movies, rock climbing or volunteering.

Inline Feedbacks
View all comments