October 2023 Summaries

Harness CI Cloud's adoption of ARM64 architecture offers significant advantages in reducing build times by up to fivefold compared to emulation, leading to substantial cost and energy savings. With ARM64, companies can achieve up to 40% better performance per dollar and reduce cloud expenses by 35%. The ARM architecture, particularly with the introduction of Apple's M1 and M2 chips, is gaining prominence in both the PC and cloud computing markets, with major service providers offering ARM solutions. Companies like Squeaky and Aerospike have reported significant reductions in cloud spending and carbon emissions after switching to ARM-based infrastructure. Despite ARM's growing momentum, some continuous integration platforms have been slow to adopt it, but Harness CI Cloud has consistently offered ARM infrastructure. The platform encourages developers to experiment with ARM64 by providing free monthly build credits and easy integration into their development pipelines.

Harness Internal Developer Portal (IDP) serves as an effective bridge between monorepos and microservices, offering features like flexible YAML definition locations, localized documentation, streamlined service onboarding, and path-aware scorecard checks, thus enhancing development workflows. While microservices have gained popularity due to the rise of Cloud Native tools, monorepos still offer significant benefits, such as simplified dependency management and efficient code reviews. The IDP is designed to accommodate both monorepos and microservices, making it versatile for organizations at different stages of their development journey. It allows YAML files to be placed anywhere in the repository for maximum flexibility, ensures documentation is closely aligned with code, facilitates the onboarding of new services without creating new repositories, and provides scorecard checks that are aware of specific paths within a monorepo. This versatility makes the Harness IDP a valuable tool for enterprises and startups looking to optimize their software development processes.

Chris DeMars recounts his experience speaking at MagnoliaJS, a developer community event in Jackson, Mississippi, characterized by its Halloween theme and a welcoming atmosphere fostered by organizers Kayla and Richard Sween. The two-day event featured a variety of talks, including Taylor Desseyn's insights on content creation and organizational scaling, Rizel Scarlett's discussion on breaking barriers with emerging tech, and Pato Vargas's expertise on React monorepos. Chris himself delivered a talk on the relationship between ROI and accessibility, highlighting his role at Harness, while also participating in fun activities like a sea captain voice-over with colleagues. The event concluded with Todd Libby's engaging talk on accessibility, underscoring the community's vibrant spirit and the value of attending or speaking at MagnoliaJS.

Harness Continuous Error Tracking (CET) is a tool designed to enhance the developer experience by providing real-time error tracking and efficient troubleshooting capabilities during the software development lifecycle. It aims to address the significant time developers spend on debugging by offering full context into errors, including code variable data, logs, stack traces, and machine data, enabling developers to detect and fix errors early. CET supports JVM-based languages, allowing developers to automate error troubleshooting, prioritize remediation efforts, and track deployment quality. It integrates with tools like JIRA and various logging and APM solutions to streamline error management and collaboration. With features such as runtime code analysis, automated root cause analysis, event management, and comprehensive dashboards, CET helps developers deliver more reliable applications faster. The tool also provides audit trails and role-based access control for compliance and security needs and is available as a SaaS or self-managed deployment.

Chargeback and showback are essential FinOps processes that enable companies to manage and hold engineering teams accountable for cloud costs by either charging them back to departmental budgets or providing visibility without actual charges. While building an in-house solution for cloud cost reporting might initially appear cost-effective, it can be resource-intensive and challenging, especially when it comes to accurately attributing shared costs and integrating advanced features like cost anomaly detection and budgeting. Harness CCM offers a comprehensive out-of-the-box solution that not only simplifies cost attribution with its Cost Categories feature but also enhances cloud cost management by providing actionable optimization recommendations, real-time anomaly detection, and advanced budgeting and forecasting capabilities. It also supports dynamic team and project management and integrates resource utilization metrics, ensuring agile and efficient cloud cost oversight. Opting for a managed solution like Harness CCM can save time and effort, allowing organizations to focus on their core business while maintaining accurate and scalable cloud cost management.

Harness Developer Hub (HDH) has transitioned from its beta phase to achieving General Availability (GA), now serving tens of thousands of unique visitors globally each month with hundreds of thousands of pageviews. Utilizing Harness for continuous integration and delivery, HDH has executed over 9000 builds with a 30% improvement in build speed, thanks to Cache Intelligence. The platform, which supports a growing number of contributors, employs a documentation-as-code approach through a Docusaurus implementation, allowing for frequent builds and deployments triggered by Git events. HDH's pipeline strategy leverages Harness capabilities, such as webhook events and cloud-based build nodes, to facilitate preview and production builds via Netlify, optimizing the process through the principles of simplicity (KISS) and eliminating redundancy (DRY). The evolution of its pipeline has included moving away from a Kubernetes-heavy approach to directly utilizing Netlify's API for preview environments, showcasing the platform's adaptability and commitment to continuous improvement in software delivery.

Switching continuous integration (CI) providers can be a complex task, but the Harness CI Migration Utility simplifies this transition by allowing users to effortlessly migrate their CI pipelines from platforms like GitLab, CircleCI, GitHub Actions, and Bitbucket to Harness. This open-source tool is particularly efficient for Drone users, as it automates much of the migration process by importing configurations and pipelines directly from Drone to Harness. The Harness CI platform, part of the Harness Modern Software Delivery Platform, offers features such as cache intelligence, test intelligence, secure cloud build machines, and extensibility through plugins to enhance build speed and security. It also provides robust governance, compliance, and shift-left security capabilities, ensuring a secure and efficient CI/CD process. The platform supports productivity with tools like the Harness AI Development Assistant and offers comprehensive dashboards for deeper insights. Harness encourages community collaboration by making the Migration Utility open-source, inviting developers to contribute to its ongoing development.

Express TypeScript is an integration of Express.js, a minimalist web application framework, with TypeScript, a statically typed superset of JavaScript, aimed at enhancing error detection, code quality, and developer productivity. This combination leverages TypeScript's static typing and rich IDE support, resulting in a more robust and streamlined development experience that reduces runtime errors and facilitates easier code maintenance. Setting up an Express TypeScript environment involves installing Node.js, Express.js, and TypeScript, along with an IDE like Visual Studio Code for efficient coding. Developers can create a simple API and build applications using this setup, benefiting from TypeScript's advanced features such as middleware usage, error handling, and testing. While there may be a learning curve for newcomers, the integration ultimately leads to improved web application development by making the code more secure and maintainable over time.

Tyler Technologies, a leading U.S. software provider for the public sector, faced escalating cloud costs due to the continuous operation of non-production environments needed for their Enterprise Permitting & Licensing solution. Initially using AWS CloudWatch to manage these costs, the company realized they needed a more strategic approach to optimize their shared cloud infrastructure. By adopting Harness Cloud AutoStopping™, Tyler Technologies was able to efficiently control idle cloud resources, initially saving $15K to $20K monthly and eventually surpassing $100K in savings. To maximize these savings, they reorganized their infrastructure by client time zone, ensuring that non-production environments were idle simultaneously to allow for more effective use of Cloud AutoStopping. This reorganization, along with insights into client activity levels, has enabled further optimization, allowing environments to be powered down for extended periods. The company's journey highlights the significant impact that thoughtful cloud architecture and management can have on cost savings, and they continue to explore additional features like Cloud Asset Governance for further efficiency.

Modern continuous delivery practices focus on automation, version control, immutability, and observability to reduce friction and cognitive overhead, allowing software tools to function smoothly for end users. These practices have roots in historical manufacturing principles, such as Taylorism and lean manufacturing, which emphasized standardization and efficiency. The evolution of software development from rigid processes like the Waterfall model to more adaptive approaches, such as Agile and continuous integration, reflects a shift toward iterative improvement and user feedback. This transformation was propelled by advances in technology, including the rise of open-source software, cloud computing, and infrastructure as code. As software delivery has become more sophisticated, the emphasis has been placed on creating a seamless user experience by treating deployment environments as immutable and ensuring that all changes are well-documented and observable. Tools and platforms like Harness are designed to simplify these complex processes, providing solutions that align with these guiding principles to enhance deployment efficiency and reliability.

Harness integrates with Terraform to manage feature flags as code, streamlining deployment pipelines and enhancing collaboration between development and operations teams. Feature flags allow toggling of application features without redeployment, supporting controlled rollouts, A/B testing, and rapid response to issues. The integration with Terraform enables precise management and automation of feature flags, offering options for advanced configurations such as permanent flags, Git integration, and rule-based targeting through Feature Flag Target Groups. This approach facilitates infrastructure as code practices, ensuring version control and traceability across environments, and empowers teams with granular control over feature releases. By using the Harness Platform Feature Flag resources, including Target Groups and Targets, organizations can automate and manage feature rollouts effectively, enhancing the accuracy and reliability of software delivery and experimentation.

Harness has been ranked first in the Gartner 2023 Critical Capabilities for DevOps Platforms report for the Platform Engineering use case, emphasizing its strengths in scaling configuration, enforcing governance without adding burden, and enhancing developer experience through templated configuration, policy as code, and multi-cloud native support. This recognition highlights the company's commitment to delivering scalable, easy-to-use, and modular tools that platform engineers appreciate. Harness's approach involves using templates to standardize best practices and policy as code techniques to balance governance with flexibility, while also offering a coherent platform that integrates with existing tools and supports multi-cloud environments. Recent advancements include Infrastructure as Code Management and an Internal Developer Portal to further aid platform engineering teams in enhancing developer effectiveness and reducing cognitive load. The company's continued focus on platform engineering aims to reduce burnout and improve productivity, ensuring a positive developer experience.

KubeCon + CloudNativeCon North America 2023, set to take place in Chicago in early November, marks the eighth year of this flagship event that brings together adopters of Cloud Native Compute Foundation (CNCF) projects. Harness, a long-time participant and contributor to the cloud-native community and Linux Foundation, will have an active presence, including booths at co-located events ArgoCon and BackstageCon. As the cloud-native landscape evolves, Harness continues to contribute to several CNCF projects, including Litmus Chaos and OpenTofu, and embraces the GitOps paradigm by contributing to Argo. The event highlights the importance of open-source and cloud technologies in achieving scale, robustness, portability, and scalability, while also noting the consolidation within the CNCF landscape, which has decreased the number of projects since 2020. Attendees are encouraged to visit Harness's booth and engage with their initiatives and offerings, underscoring their commitment to supporting and advancing cloud-native technologies.

Feature flag-driven development, a technique that enables developers to toggle features on and off without full deployments, is revolutionizing software development by enhancing agility, control, and speed. This approach allows for gradual feature releases, real-time experimentation, and risk mitigation, enabling teams to innovate continuously while delivering optimized user experiences. Tools like Split by Harness facilitate this process, offering a feature management dashboard for creating, managing, and monitoring feature flags, and providing insights into feature performance and user engagement. Feature flags integrate seamlessly with CI/CD pipelines, enabling continuous delivery and reducing technical debt by decoupling code deployment from feature activation. They also support A/B testing, canary releases, trunk-based development, and user personalization, allowing for targeted testing, quick problem-solving, and minimizing disruptions through quick rollbacks. Harness’s Feature Management & Experimentation further enhances this approach with capabilities for precise audience targeting, robust metrics tracking, collaboration, security, and real-time feedback, setting it apart from competitors like LaunchDarkly by providing 100% feature-level observability. Embracing feature flag-driven development with tools like Harness empowers development teams to innovate rapidly, make data-driven decisions, and deliver cutting-edge applications that align with user needs and expectations.

Harness has implemented a range of updates across its products to enhance user experience and streamline processes in continuous delivery, integration, and cloud cost management. These updates include improved UI feedback for approval steps, enhanced service propagation for multi-service stages, and new regex filter support for artifact paths in the continuous delivery segment. Continuous integration updates introduce new S3 step support, error analysis enhancements, and expanded test intelligence capabilities. Cloud cost management now offers better visualization of negative costs and the ability to copy cost category buckets. Service reliability management has added composite SLO support and new filtering options, while chaos engineering introduces new protocol support and targeting capabilities. The internal developer portal has enhanced plugin support and introduced scorecards for better data insights. Platform improvements include extended delegate expiration, optimized ACL cleanup, and UI enhancements for roles and authentication settings. The self-managed enterprise edition now supports self-managed MinIO object storage for disaster recovery and improved Redis database support, among other enhancements. Early access features offer rollback support for specific deployment types and new dashboard functionalities.

GitOps is likened to constructing a skyscraper, requiring a phased approach of experimentation, foundation building, and scaling, with challenges and solutions evolving at each stage. The early phase, Day 0, involves familiarizing with GitOps fundamentals and setting up basic workflows, while Day 1 focuses on separating application code from configurations and avoiding complications in continuous integration pipelines. Day 2 presents automation complexities, such as the bootstrapping dilemma and tool reliability, where Terraform emerges as a key solution by providing infrastructure as code to ensure consistency and define clear end goals. Terraform's idempotency and ability to handle complex setups make it invaluable for scaling GitOps effectively. Dewan Ahmed, a seasoned developer advocate, suggests that approaching software development with the same meticulous strategy as building skyscrapers can lead to strong, scalable, and impressive outcomes, advocating for the use of tools like the Harness Terraform Provider to enhance GitOps efficiency.

Harness has introduced an AI-powered customer support tool, Harness Support powered by Harness AI, to enhance the customer experience by providing quick and accurate responses to queries about Harness products and documentation. This tool integrates seamlessly into the product, using advanced AI algorithms and GenAI techniques to transform user queries into actionable insights, offering real-time answers and direct links to relevant resources. By employing a vector database, the AI provides contextually relevant and precise answers, significantly improving the efficiency of finding information without the need for traditional support tickets. This innovation reflects Harness's commitment to leveraging AI technology to streamline interactions and ensure users have immediate access to the information they need, effectively transforming customer support in the fast-paced digital age.

Elevating product development processes through statistical rigor is crucial, as illustrated by a dining scenario where choosing among taco restaurants highlights the importance of systematic analysis over subjective opinions or simplistic data comparisons. The author emphasizes that relying on small data points or averages can lead to misleading decisions, which, while inconsequential for choosing a restaurant, can have significant repercussions in product development. A rigorous statistical approach, akin to the third method discussed, involves a comprehensive analysis of available data to make informed decisions, avoiding reliance on singular opinions or inadequate data samples. Split's Feature Data Platform enables this approach by offering tools for safe feature deployment and experimentation without requiring advanced statistical knowledge, ensuring that product decisions are data-driven and robust. The platform facilitates feature flagging, A/B testing, and contextual data analysis, helping teams move quickly and confidently in their development efforts.

Deploying applications to Kubernetes using Gitness involves setting up Gitness as a containerized application, configuring Kubernetes clusters, and creating deployment pipelines that leverage Gitness's continuous integration and deployment capabilities. Gitness, an open-source Git solution, offers reliable Git repository hosting along with capabilities for code hosting, collaboration, and continuous integration, making it a comprehensive software delivery platform. Users can create projects and repositories within Gitness to manage their code and deployment processes. The blog highlights the setup process for Gitness, including creating service accounts and configuring secrets for Kubernetes deployment, and demonstrates the creation of a deployment pipeline that uses shell scripts to automate the deployment of an nginx application. Although Gitness is still evolving and faces some challenges, such as the exposure of secrets in pipeline logs and reliance on shell scripts, it aims to streamline the DevOps cycle and set new standards for open-source code hosting. The author, Dewan Ahmed, provides insights into the current state and future potential of Gitness while also sharing his expertise in developer advocacy and infrastructure solutions.

Harness effectively utilizes feature flags to manage feature releases, providing early access, rollbacks, gradual rollouts, and customer-specific troubleshooting to enhance user experience and minimize disruptions. Before launching Harness Feature Flags, internal teams stress-tested the product, and it continues to be an integral tool for teams like Customer Success Management (CSM) for various purposes. The CSM team uses feature flags to offer select customers early access to new features, gather feedback, and ensure smooth rollouts by mitigating issues through quick rollbacks if necessary. Additionally, feature flags enable gradual rollouts to monitor performance, customer-specific troubleshooting to tailor solutions, and testing in non-production environments to minimize risks. Harness Feature Flags also support automated rollout pipelines, governance with robust policy tools, automated lifecycle management to handle tech debt, and integration with CI/CD for a seamless software development lifecycle.

Harness feature flags are a tool that enables safe testing of new software features in production by allowing developers to target specific user groups, gradually roll out changes, and quickly revert to previous versions if necessary. This approach reduces risks, accelerates development, and enhances software quality by identifying bugs in real-world conditions and gathering early user feedback. By using feature flags, companies can test features like new algorithms or app functionalities with a limited user base before a full-scale release, ensuring performance issues are addressed beforehand. Testing in production, while initially daunting, is beneficial for catching hard-to-find bugs and aligning software performance with real-world user expectations, ultimately leading to a faster and more reliable development cycle.