Abstract: Methods, systems, apparatus including computer programming encoded on a computer storage medium for efficiently storing data across memory hierarchy on one or plurality of nodes include creating ordered partitions of data based on key where partition order is achieved using highly space optimized prefix tree index. Within a partition, data is fully or partially ordered based on key. Multiple prefix tree indices are created one for each memory hierarchy and all are stored in fast memory like DRAM. Data is merged or moved from faster memory to slower memory in the hierarchy as the space is used up. Checkpoint mechanism along with WAL provides recovery guarantee and data snapshots. Distributed data storage systems like databases/key value stores can utilize this data storage mechanism to store and retrieve data efficiently across memory hierarchy on one or plurality of nodes.

Abstract: Methods, systems, apparatus including computer programming encoded on a computer storage medium for efficiently storing data across memory hierarchy on one or plurality of nodes include creating ordered partitions of data based on key where partition order is achieved using highly space optimized prefix tree index. Within a partition, data is fully or partially ordered based on key. Multiple prefix tree indices are created one for each memory hierarchy and all are stored in fast memory like DRAM. Data is merged or moved from faster memory to slower memory in the hierarchy as the space is used up. Checkpoint mechanism along with WAL provides recovery guarantee and data snapshots. Distributed data storage systems like databases/key value stores can utilize this data storage mechanism to store and retrieve data efficiently across memory hierarchy on one or plurality of nodes.

Abstract: An embodiment includes a system, comprising: a server coupled to a plurality of nodes and configured to: select a node from among the nodes to perform a maintenance operation; instruct the selected node to perform the maintenance operation; and respond to access requests based on the selected node; wherein performing the maintenance operation by the selected node decreases a performance of the selected node.

Abstract: A replication manager for a distributed storage system comprises an input/output (I/O) interface, a device characteristics sorter, a routing table reorderer and a read-query load balancer. The I/O interface receives device-characteristics information for each persistent storage device of a plurality of persistent storage devices in which one or more replicas of data are stored on the plurality of persistent storage devices. The device characteristics sorter sorts the device-characteristics information based on a free block count for each persistent storage device. The routing table reorderer reorders an ordering of the replicas on the plurality of persistent storage devices based on the free block count for each persistent storage device, and the read-query load balancer selects a replica for a received read query by routing the received read query to a location of the selected replica based the ordering of the replicas stored on the plurality of persistent storage devices.