/tag/kubernetes

  • Microservice Deployments

    Kubernetes is a container orchestrator for both short-running (such as workflow/pipeline stages) jobs and long-running (such as web and database servers) services. Containerized applications running in the UVARC Kubernetes cluster are visible to UVA Research networks (and therefore from Rivanna, Skyline, etc.). Web applications can be made visible to the UVA campus or the public Internet. Kubernetes Research Computing runs microservices in a Kubernetes cluster that automates the deployment of many containers, making their management easy and scalable. This cluster will eventually consist of several dozen instances, >2000 cores and >2TB of memory allocated to running containerized services. It will also have over 300TB of cluster storage and can attach to both project and standard storage.
  • Microservices

    – Microservice architecture is an approach to designing and running applications as a distributed set of components or layers. Such applications are typically run within containers, made popular in the last few years by Docker. Containers are portable, efficient, reusable, and contain code and any dependencies in a single package. Containerized services typically run a single process, rather than an entire stack within the same environment. This allows developers to replace, scale, or troubleshoot portions of their entire application at a time. General Availability (GA) of Kubernetes - Research Computing now manages microservice orchestration with Kubernetes, the open-source tool from Google.
  • Computing Environments at UVA

    Rivanna The primary vehicle for high-performance computing since 2014 has been the Rivanna cluster. Rivanna is a heterogenous system with a total of 603 nodes and 20476 cpu cores. It consists of 527 nodes with 20-40 cores and 128-768GB of RAM each, 11 large memory nodes with 16-48 cores and 1-1.5TB of RAM each, and 29 nodes with a total of 172 NVIDIA GPU accelerators (K80, P100, V100, A100, RTX-2080 Ti). These nodes are partitioned for various types of workloads, but include development, parallel, HTC and instructional partitions. All nodes are supported by a high-performance EDR/FDR Infiniband network using Mellanox hardware.