BisQue has now moved its entire orchestration system to align with the latest software trends for distributed systems. With the move to Kubernetes, BisQue is now highly available (fault tolerant) and scalable across multiple nodes. We achieved this by taking advantage of the Kubernetes control loop that constantly checks the defined state of our deployment. If any of our pods or nodes fail, resources are rescheduled automatically to match BisQue’s desired state. GPU nodes are now labeled and can have jobs that require GPU resources scheduled on those nodes without any manual interaction. This move also allows for a larger community adoption since users can install BisQue on a local K3s, K3D, or full K8s on-premise cluster or on a cloud provider (GCP, AWS, etc.). We will be releasing Helm charts, which are fully packaged BisQue deployments that can be deployed on any Kubernetes cluster
