June 2020 Summaries

Harness has introduced Cloud Cost Management to provide engineering teams with detailed insights into cloud expenses, allowing proactive cost-saving measures without needing to tag resources. This system provides real-time visibility into cloud spending, alleviating the unpredictability of cloud costs. Enhancements like Git Sync Activity offer a granular audit of configuration changes, facilitating better debugging and configuration management, while a new feature allows exporting deployment logs for compliance purposes. Additionally, Harness has been actively engaging the community through events like DevOps Zooms, where they discuss building confidence and consistency in the cloud. The company has also received notable media attention for its efforts in helping DevOps teams manage cloud costs effectively.

Database migrations pose significant risks, particularly regarding data loss and compatibility issues, which can be mitigated using feature flags to enable safe, incremental updates. Feature flags allow developers to decouple schema changes from feature rollouts, providing a mechanism to test and revert changes without risking data integrity. The text highlights the importance of the expand-contract pattern, which ensures that different versions of software and database schemas can coexist, allowing gradual changes while maintaining backward compatibility. By implementing feature flags, teams can manage schema updates with reduced risk, ensuring that any code changes are independent of database modifications. This process not only safeguards against potential bugs during updates but also enhances the confidence to move quickly without compromising system stability. The Split Feature Data Platform is presented as a tool that facilitates the safe deployment of features, allowing for controlled rollouts and feature experiments, thereby revolutionizing how development work is conducted.

Harness Cloud Cost Management is a cutting-edge financial operations tool that has enabled users to achieve significant cost savings and improved visibility into cloud expenses, fostering a culture of cost-awareness and continuous optimization. The tool empowers engineering teams by providing actionable cost insights traditionally reserved for financial analysts, allowing for better resource allocation and cost transparency. Users have reported substantial savings, such as over $500,000 annually, by employing practices like root-cost analysis and granular cost governance, which help identify inefficiencies in resource usage. The platform allows for cost allocations by application, service, or cluster, eliminating the need for meticulous tagging and enabling organizations to understand the financial implications of their operations more effectively. With features that include hourly cost granularity and proactive alerts, Harness ensures continuous monitoring and adjustment of cloud spending, emphasizing that cost management is an ongoing process rather than a one-time fix.

Harness Cloud Cost Management empowers engineering teams with self-service tools to manage and optimize cloud expenditures effectively, leading to significant cost reductions, such as the 40% decrease in annual cloud costs achieved by Harness itself. This approach shifts the responsibility of cloud cost management from dedicated finance or operations teams to the engineers who consume cloud resources, thus eliminating inefficiencies and waste. The platform provides detailed visibility into cloud costs by offering features like root cost analysis for Kubernetes and AWS ECS clusters, hourly time granularity for cost tracking, and integration with continuous delivery systems. By correlating cloud events with cost spikes and providing insights into resource utilization, engineers can proactively adjust their resource allocations and reduce unnecessary expenditures. Additionally, Harness Cloud Cost Management supports multi-cloud environments and offers standard features such as executive dashboards, budget alerts, and historical reporting, making it a comprehensive tool for cloud financial optimization.

Java EE, built on Java SE, is a specification for enterprise Java extensions that is often misunderstood as heavyweight, despite being designed for lightweight performance. The Java EE 8 release, facing delays due to Oracle's shifting priorities, has sparked community-driven initiatives like the Java EE Guardians advocating for open standards and collaboration. Java EE differs from Java SE by providing enterprise-level extensions and is not a standalone implementation but a specification with implementations from vendors like Oracle, RedHat, and IBM. Challenges include licensing issues and delays caused by Oracle's focus on cloud services, leading to concerns about the future of Java EE. Despite these issues, Java EE remains a crucial part of the Java ecosystem with many committed contributors. The Java EE Guardians, founded by Reza Rahman, aim to ensure the community takes a more active role in its development, highlighting the need for broader community involvement to compensate for Oracle's diminished activity.

Harness has strengthened its engagement in the DevOps community through various initiatives, including joining the Continuous Delivery Foundation and DevOps Institute, launching Harness University, and expanding its online presence with increased community page views and new content. The company has introduced several ambassador programs, with Tiffany Jachja joining both the Continuous Delivery Foundation and the DevOps Institute as an ambassador, reflecting a commitment to fostering collaboration and learning. Harness has also sponsored events like the DevOps Institute's SKILup Day and launched platforms such as "Deliver Better" to share learning content and stories across different roles within enterprise software delivery. The organization is focusing on addressing interoperability challenges in CI/CD tools and technologies by contributing to the CD Foundation's Interoperability Special Interest Group. Harness University continues to offer educational resources, such as mastering Kubernetes Canary Deployments, while the company's "Harness and the Masters of Continuous Delivery" program aims to highlight insights from active users of its CD platform. With these efforts, Harness aims to continue supporting and celebrating its community throughout the summer and beyond.

A/B testing and multivariate testing are two methods used to optimize websites and features, each with distinct characteristics and applications. A/B testing, or split testing, involves comparing two versions of a webpage, feature, or user flow to determine which performs better based on metrics like conversion rates, and is typically quicker due to its simplicity. Multivariate testing, on the other hand, is more complex, involving multiple variables simultaneously to gain deeper insights into which combinations of elements improve a webpage's performance, making it suitable for high-traffic platforms. While A/B testing is ideal for lower-traffic sites due to its faster results with fewer variations, multivariate testing requires a substantial amount of traffic to achieve statistically significant results. Split, a feature management platform, facilitates these testing methods by allowing users to set up feature flags, control feature deployment, and conduct feature experiments seamlessly, enhancing the efficiency and reliability of software releases.