Nscale Platform Services
Product · SaaS / cloud software · Engineering & development by Nscale
Production-grade orchestration for AI workloads with Kubernetes and Slurm.
Nscale Platform Services provide production-grade orchestration and management tools for AI workloads, combining Kubernetes and Slurm scheduling. The Nscale Kubernetes Service (NKS) allows teams to provision isolated clusters for containerized workloads in minutes, while Managed Slurm offers an HPC-grade batch scheduler for large-scale training runs. The platform supports mixed workloads, allowing batch training and interactive jobs to coexist with predictable scheduling. Users can also manage virtual machines, VPC networks, and shared persistent storage through the Nscale Console or via Terra
- Languages
- English
- Orchestration Engines
- Kubernetes, Slurm
- Storage Type
- Shared persistent NFS storage
- Access Method
- Nscale Console, SSH, Terraform Provider
- Isolation Method
- VPC Networks, RBAC, Multi-tenancy
Overview
Nscale Platform Services provide production-grade orchestration and management tools for AI workloads, combining Kubernetes and Slurm scheduling. The Nscale Kubernetes Service (NKS) allows teams to provision isolated clusters for containerized workloads in minutes, while Managed Slurm offers an HPC-grade batch scheduler for large-scale training runs. The platform supports mixed workloads, allowing batch training and interactive jobs to coexist with predictable scheduling. Users can also manage virtual machines, VPC networks, and shared persistent storage through the Nscale Console or via Terraform. This suite is designed to reduce operational bottlenecks and accelerate developer velocity for AI-native companies.
Nscale Kubernetes Service (NKS)
- Production Kubernetes: Hardened orchestration service designed for production workloads with integrated observability and networking.
- Lightweight Clusters: Fast spin-up and multitenancy isolation for development and test scenarios.
- Rapid Provisioning: Provision isolated Kubernetes clusters for container workloads in minutes.
Managed Slurm
- Batch Scheduling: HPC-grade Slurm batch scheduler that runs on Kubernetes for managing large GPU training runs.
- Mixed Workload Support: Simplify management of mixed workloads across containerized services and batch training.
Platform Infrastructure
- Virtual Machines: Create GPU and CPU virtual machines with SSH access for experimental workloads.
- VPC Networks: Isolated private networks for resources to ensure data and network sovereignty.
- Shared Filesystem: Shared persistent NFS storage for instances and clusters.
- Terraform Provider: Manage Nscale infrastructure as code using the official Terraform provider.
Specifications
- Orchestration Engines
- Kubernetes, Slurm
- Storage Type
- Shared persistent NFS storage
- Access Method
- Nscale Console, SSH, Terraform Provider
- Isolation Method
- VPC Networks, RBAC, Multi-tenancy
Added by AntSpace from public sources (nscale.com).