Improving DevOps Efficiencies by Building an Agile Data Tier
August 10, 2016

Michelle McLean
ScaleArc

The ability to react quickly to market changes and launch new features and functionality is critical to maintaining a competitive edge.
 
Yet today, DevOps teams must spend substantial time and resources dealing with database scalability. If they need to consider scalability implications every time they roll out new code, the agile process breaks down. Software rollout cycles can get bogged down in application complexity that requires applications to be changed to function with the various roles of database servers and to regulate traffic to ensure data being read from secondary servers is current; database provisioning requirements which can be time-consuming and inefficient, limiting agility; and a lack of visibility – not understanding what application behaviors cause databases to slow down, which increases the development time required to address those issues.
 
Using database load balancing software, DevOps teams can solve these problems by creating an agile data tier that guarantees scalability of their database stack. An agile data tier can slash development time from months to just weeks by eliminating the need to build database scalability functionality into each separate application, enabling database clusters to be easily built and shared at scale and ensuring visibility into database traffic.

Enable Automated Database Failover and Scalability Without Increasing App Complexity

To scale structured data, DevOps teams follow a very specific set of rules. Functionality such as replication-aware load balancing, read/write split, and caching must be programmed into applications for them to handle and take advantage of multiple database servers. Applications also need to understand how to fail over to the right server within a cluster for high availability (HA).

A lot of code must be created within an app for it to be able to handle high availability (HA) –taking up valuable development time. Likewise, building an app that understands database topology – to do read/write split and assist in automated failover – adds to application complexity.

While typical caching solutions – such as Memcached – can increase database speed and scalability, they add complexity at the application layer. With these solutions, developers must spend time writing code that can create, manage, and expire cache in NoSQL. They must also write the individual business rules that specify which type of data can be cached. Similarly, a lot of code must be created within an app for it to be able to handle high availability (HA) – again, taking up valuable development time.

Remove the Need for Application Modification

Database load balancing software transparently provides replication-aware load balancing and read/write split to any application. By monitoring the replication lag between primary and secondary servers, database load balancing software can intelligently distribute queries, making sure an application gets the freshest data from the secondary servers. DevOps teams can easily set the appropriate replication lag threshold for a particular application.

Database load balancing software also allows teams to analyze database traffic and find read queries that are good candidates for caching. They can then choose which queries they want to cache and for how long. The software will then cache query responses and serve them to the application from a transparent in-memory cache without any changes at the application layer.

DevOps teams can also enable automated failover – without changing a single thing at the application layer. The database load balancing software understands which queries are reads and which are writes/transactions. It load balances the reads across all available servers and routes the writes to the primary. If the primary fails, the database load balancing software can enable seamless cross-server or cross-data center failover, maintaining state and reducing app errors with no changes to your applications.

By removing database scalability from the application and using database load balancing software, you can:

■ Roll out applications faster, since they are simpler to develop

■ Scale  database stack without requiring any new application coding

■ Focus staff attention on adding features and functionality that serve the business, instead of programming for data distribution

Provision Databases Seamlessly – Internal DBaaS

With database load balancing software, they can easily carve up a large database cluster into smaller pieces, giving each user the ability to run a certain amount of concurrent load on that database cluster. This approach provides a regulated, controlled load management stack – similar to a resource governor. They can slice and dice a cluster into smaller pieces, giving users access to their own sets of resources without disturbing anyone else on the cluster, essentially creating an internal DBaaS for their organization.

In this architecture, provisioning a new database endpoint takes only a few seconds compared with minutes or even hours when using an individual VM for each developer – similar to Amazon’s RDS that essentially provides a database ‘instance’ as a service.

With an internal DBaaS, teams can provide their application developers with an already highly redundant environment, ensuring that developers and QA are not interrupted due to database downtime. Teams also get access to very deep SQL analytics that help them test their applications better and identify potential application problems before moving into production.

Build New Applications, Scale and Solve Problems Faster and More Effectively

By decreasing application complexity, database load balancing software lets DevOps teams scale more effectively, benefit from DBaaS, and get the control and visibility needed to achieve results faster. Because database scalability functions are now out of the application, application design is simpler, making it easier to develop and build new applications. DevOps teams effortlessly provision databases to test your application’s functionality and benchmark performance better, with greater visibility into what the application is doing with the database.

With increased application resiliency, better use of database resources and enhanced system performance, organizations can speed the development process, improve overall efficiencies and gain business advantage.

Michelle McLean is VP of Marketing at ScaleArc.

Share this

Industry News

July 13, 2020

Docker announced a collaboration with Amazon Web Services (AWS) to simplify the lives of developers by allowing them to focus on application development, streamlining the process of deploying and managing containers in AWS from their local development environment.

July 13, 2020

Perforce Software announced the release of a combined JRebel and XRebel plugin for the Eclipse IDE.

July 13, 2020

Spectro Cloud announced that its first product—Spectro Cloud—is now generally available.

July 09, 2020

ShiftLeft released a new version of NextGen Static Analysis (NG SAST), including new workflows, purpose-built for developers that significantly improve security, while enhancing productivity.

July 09, 2020

RunSafe Security announced a partnership with JFrog that will enable RunSafe to supercharge binary protections via a simple plugin that JFrog users can deploy within their Artifactory repositories and instantly protect binaries and containers.

July 09, 2020

LeanIX closed $80 million in Series D funding led by new investor Goldman Sachs Growth.

July 08, 2020

Afi.ai introduced Afi Data Platform, a cloud-based replication and resiliency service that helps to monitor, predict downtime and recover K8s applications.

July 08, 2020

D2iQ announced the release of Conductor, a new interactive learning platform that enables enterprises to access hands-on cloud native courses and training.

July 08, 2020

SUSE entered into a definitive agreement to acquire Rancher Labs.

July 07, 2020

Micro Focus announced AI-powered enhancements to the intelligent testing capabilities of the UFT Family, a unified set of solutions designed to reduce the overall complexity of automating the functional testing processes.

July 07, 2020

Push Technology announced the launch of a new Service API capability for Diffusion Cloud, Push’s Real-Time API Management Cloud Platform.

July 07, 2020

Lightrun exited stealth and announced $4M in seed funding for the first complete continuous debugging and observability platform for production applications.

July 01, 2020

JFrog announced the launch of ChartCenter, a free, security-focused central repository of Helm charts for the community.

July 01, 2020

Kong announced a significant upgrade to open source Kuma, Kuma 0.6, available today.

July 01, 2020

Compuware Corporation, a BMC company, announced new capabilities that further automate and integrate test data and test case execution, empowering IT teams to achieve high-performance application development quality, velocity and efficiency.