![]() ![]() Provisioning a volume might result in a non-empty volume: Not met by the normal Kubernetes EmptyDir volumes. Request local storage with performance characteristics or size that is Local LVM storage as scratch spaceĪpplications working with data sets that exceed the RAM size can When deploying memcached through one of the appĬontrollers, generic ephemeral volumes make it possible to request a PMEM volume Recent releases of memcached added support for using Persistent ![]() Use cases: Use cases Persistent Memory as DRAM replacement for memcached This makes generic ephemeral volumes a suitable solution for several No need anymore to implement and configure scheduler extenders and The Kubernetes scheduler itself picks suitable nodes, i.e.The volumes are usable with any app controller that accepts.Resizing, the future storage capacity tracking, etc.) All of the typical volume operations (snapshotting,.Volumes may have some initial data, depending on the driver and.Volumes and (for capacity tracking) implements the CSI GetCapacity call. Works with any CSI driver that supports provisioning of persistent.Volumes can have a fixed size that applications are never able to exceed.Storage can be local or network-attached.Volumes that are conceptually more like the EmptyDir volumes: Therefore, Kubernetes 1.19 introduces two new alpha features for This is problematic for volumes which consume significant resources onĪ node or for special storage that is only available on some nodes. normal, standard-compliant CSI drivers will not work, andīy design such volumes are supposed to be usable on whatever node These injectĪrbitrary states, such as configuration, secrets, identity, variablesĬSI drivers must be modified to support this Kubernetes feature, Made it possible to extend Kubernetes with CSIĭrivers that provide light-weight, local volumes. Kubernetes already supports several kinds of such ephemeralįunctionality of those is limited to what is implemented inside Other applications expect some read-only inputĭata to be present in files, like configuration data or secret keys. Used data into storage that is slower than memory with little impact Services are often limited by memory size and can move infrequently Some applications need additional storage but don't care whether thatĭata is stored persistently across restarts.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |