Modernizing HackerRank’s build release pipeline and optimizing AWS infrastructure
Meet HackerRank
HackerRank, founded in California in 2012, enables more than 3,000 companies worldwide to hire skilled developers. With HackerRank’s proprietary platform, employers can screen and evaluate candidates’ programming skills through the use of coding challenges, assessments, and technical reviews. The platform enables tech recruiters and hiring managers to objectively evaluate talent at every stage of the recruiting process, from candidate screening to job offers.
HackerRank also offers a community platform where developers can share knowledge, showcase their skills, solve coding challenges, participate in coding contests, and collaborate with other developers worldwide.
The Challenge
As HackerRank expanded, its user base grew rapidly, increasing the need for more robust product capabilities. To evolve its product portfolio with agility, the engineering team needed to optimize productivity without driving up expenses in its AWS infrastructure. One of the biggest challenges for both productivity and cost was downtime, which caused significant frustration for HackerRank engineers and customers. HackerRank turned to DoiT for cloud consulting and infrastructure optimization.
As a DoiT customer, HackerRank has access to more than 200 engineers, including cloud architects, cloud engineers, and FinOps experts. As soon as DoiT identified the issue, the team moved quickly to address pipeline inefficiencies and launched a Pipeline Modernization project with HackerRank.
The Solution
After several technical deep dives, the DoiT team identified the root causes: the current build-release pipeline introduced delays and lengthened release cycles, creating inefficiencies. As a result, DoiT recommended a multi-step approach. The plan included optimizing Docker container startup time and providing the HackerRank team with code samples and documentation for a proof of concept (PoC). Additionally, DoiT planned to run unit tests in parallel and move the build server to Amazon Elastic Container Service (ECS) with AWS Fargate, providing greater elasticity and lower maintenance.
The DoiT FinOps team initiated an engagement to reduce AWS infrastructure costs. While HackerRank had already purchased Savings Plans (SPs), it decided to leverage DoiT’s support and PerfectScale for Spot in addition to these SPs.
Using machine learning, DoiT’s technology continuously monitors HackerRank’s cloud usage to identify compute usage not already covered by commitments. This increases savings without requiring additional 1-year Savings Plans or Reserved Instances or the forecasting and ongoing management they entail.
Additionally, PerfectScale for Spot helped HackerRank identify an opportunity to optimize the costs of the Auto Scaling groups (ASGs) that organize its EKS worker nodes in non-production environments. By managing the instance composition of its ASGs, PerfectScale for Spot helped HackerRank adopt Spot Instances and save on on-demand compute costs while reducing downtime risk.
PerfectScale for Spot made best-practice recommendations to its ASG configurations to support more instance types with similar attributes to the originally selected type. In addition, PerfectScale for Spot has a “fallback to on-demand” feature, which automatically uses on-demand instances when there isn’t enough Spot capacity for its designated instance types. This enabled HackerRank to confidently modify many of its non-production ASGs to use Spot Instances as much as possible.
The Results
HackerRank partners with over 2,000 organizations, from startups to Fortune 500 companies, and is expanding into other fields such as skills certification, education, and enterprise software solutions.
Thanks to the Pipeline Modernization project, HackerRank has streamlined its build-release cycle and reduced overhead, allowing engineers to release small changes more frequently and with greater agility. y aligning more closely with the Agile practice of incremental delivery, HackerRank now provides faster, more reliable service to its customers.
With DoiT’s technology, HackerRank decreased on-demand EC2 expenses by 24%, saving thousands of dollars without extra effort or risk. Furthermore, through increased utilization of AWS Spot Instances using PerfectScale for Spot, HackerRank cut the on-demand EC2 spend of its non-production ASGs by nearly 50%. DoiT provided intelligent technology to simplify and automate cloud use, alongside focused, expert consulting and technical support.