4th workshop on advances in open runtimes and cloud performance technologies.

Cloud services such as IBM Cloud or Amazon Web Services are increasingly becoming the environments where applications are developed, tested, and deployed, data gets stored, and businesses are run. Many of the features that define a cloud (e.g., resiliency, elasticity, consistency, security) are realized through runtime technologies. Clouds are polyglot environments, and therefore advances in cloud development are directly driven by innovation in runtime technologies. However, cloud environments pose unique and often conflicting demands on runtime systems that are generally less of a concern in isolated systems. Throughput performance (how many results can my application produce?), density (how many instances of my application can I create and run simultaneously in my provisioned environment?), startup performance (how quickly can I start a new instance of my application?), and language interoperability are all examples of important considerations that require innovative solutions. Modern language runtimes are complex, dynamic systems that involve a myriad of components that must work cooperatively to achieve the functional and performance requirements of a given language. Typical core runtime technologies include dynamic just-in-time compilers for performance, garbage collection for heap management, platform abstraction for ease of portability to different hardware and operating system environments, test infrastructure for quality control, developer tooling for diagnosis and tuning of the various components, and interoperability between different language environments. Cloud workloads are typically containerized and employ microservice and serverless architectures. Achieving peak performance in such environments requires careful tuning of the cloud services and applications in concert with the runtime system. The goal of this workshop was to bring together research, industry, and developers from runtimes and cloud communities to share and discuss innovations, challenges, and research across a broad set of open source technologies (such as Eclipse OMR, Eclipse OpenJ9, Node.js, Open Liberty, Kubernetes) to improve performance in cloud environments. The focus on open solutions rather than proprietary was key as it allowed for greater collaboration amongst individuals, communities, researchers, and industry through shared learning on common technology. This was a full-day virtual speaker session workshop where researchers, students, and practitioners presented their recent innovative work and findings. Topics discussed in the workshop included, but were not limited to: • Open runtime technology frameworks; • Compiler technology and innovative optimizations for dynamic cloud environments; • Garbage collection and memory subsystem performance; • Runtimes/Cloud cooperative tuning; • Hardware techniques to assist runtime technologies; • Dynamic languages for the cloud; • Testing and correctness of runtime technology; • Throughput and startup performance, and memory footprint reduction; • Use of tools and infrastructure built on open technologies; and • Innovative ways of exploiting open runtime technologies. This workshop was successful in exploring a wide range of innovative runtime problems and solutions for the cloud environments. Many of the innovations were based on the open-source Eclipse OMR and Eclipse OpenJ9 projects. Eclipse OMR is a toolkit of language-agnostic runtime components that can be integrated in runtime environments to provide or extend the desired runtime features. The most popular components include garbage collection and compilation technologies as well as a common, portable interface for abstracting operating system functionality. Eclipse OpenJ9 is an open source, high-performance Java Virtual Machine that fully implements that Java Virtual Machine Specification and is used by several open source Java projects such as Open Liberty.