About IBM Storage Scale with IBM Spectrum LSF

The default shared file storage solution for your IBM Spectrum LSF cluster is IBM Cloud® File Storage for VPC, which can connect up to 256 hosts per zone per VPC. This limits the maximum number of dynamic compute nodes to 250. Instead, you can use IBM® Storage Scale as your storage solution, which can mount up to 4000 NFS connections per protocol and can be extended to a 32-node protocol cluster. If you use Storage Scale as your storage solution, you first set up a Storage Scale cluster, and then integrate a list of values from the Storage Scale deployment with the IBM Spectrum LSF cluster deployment.

Storage Scale is a clustered file system that provides concurrent access to a single file system or set of file systems from multiple nodes. It enables high-performance access to a common set of data to support a scale-out solution or to provide a high availability platform. Storage Scale can run on virtualized instances that provide common data access in environments, and uses logical partitioning or other hypervisors. Multiple Storage Scale clusters can share data within a location or across a wide area network (WAN) connections.

Storage Scale is a file system that is defined over one or more nodes. On each node in the cluster, Storage Scale includes three components: administration commands, a kernel extension, and a multithreaded daemon.

Also, you can use Storage Scale's single system to share dedicated file systems for data storage. All of your IBM Cloud dynamic nodes can access the shared data throughout the lifecycle of the cluster through Cluster Export Services (CES) NFS mount points. CES is part of the IBM Storage Scale architecture, and enables access to your data. When you deploy an IBM Storage Scale cluster, you also configure CES deployment values so that Storage Scale and IBM Spectrum LSF can share data.

For more information, see the Storage Scale documentation.