Abstract: Creating replicas using queries may be implemented for a time series database. A new host for a new copy of time series database data may be added and idempotent ingestion of additional data to be included in the new copy after a creation time for the new copy may be performed. Queries to other hosts that store the time series database data may be performed to obtain time series data prior to the creation time. Idempotent ingestion of the results of the queries may be performed at the new host after which performance of queries to the new copy of the time series database may be allowed at the new host.

Abstract: Methods, systems, and computer-readable media for schema-based spatial partitioning in a time-series database are disclosed. A time-series database divides elements of time-series data of a plurality of time series into partitions according to a clustering scheme. The time series are associated with respective schemas. The schemas comprise dimension names and measure names. The clustering scheme is determined based (at least in part) on the schemas and dimension values, and the schemas are determined based (at least in part) on the elements of time-series data. The time-series database stores the elements of time-series data from the plurality of partitions into one or more storage tiers. The time-series database performs a query of the time-series data in one or more of the storage tiers. The query is performed based (at least in part) on the clustering scheme.

Abstract: Methods, systems, and computer-readable media for a scalable metadata index for a time-series database are disclosed. A time-series database receives a query comprising spatial and temporal boundaries of requested time-series data in a table. Using traversal of a metadata index, the time-series database determines one or more storage locations in the one or more data sources corresponding to the spatial and temporal boundaries. The metadata index includes a root node representing spatial and temporal boundaries of the table. The metadata index includes one or more leaf nodes representing the spatial and temporal boundaries of the requested time-series data. The one or more leaf nodes indicate the one or more storage locations in the one or more data sources. The time-series database obtains one or more elements of the requested time-series data from the one or more storage locations in the one or more data sources.

Abstract: Methods, systems, and computer-readable media for auto-scaling using temporal splits in a time-series database are disclosed. One or more stream processors of a time-series database write time-series data of a plurality of time series into a plurality of two-dimensional tiles, including an individual tile representing spatial boundaries and temporal boundaries. The heat of time-series data written to the individual tile exceeds a threshold. The current time is compared to the temporal boundaries, and the comparison indicates that the current time is beyond a threshold point within the temporal boundaries. Based at least in part on the comparison, a split is performed of the individual tile into a first new tile and a second new tile. The first new tile represents the spatial boundaries and a first portion of the temporal boundaries, and the second new tile represents the spatial boundaries and a second portion of the temporal boundaries.

Abstract: Methods, systems, and computer-readable media for replica group modification in a distributed database are disclosed. One or more stream processors write elements of a data stream to a plurality of replicas in a replica group. A control plane adds an identifier of a new replica to membership metadata for the replica group. The control plane also generates a new version number for the membership metadata that now includes the new replica. The stream processor(s) write new elements of the data stream to the replica group including the new replica. Write requests to the replica group comprise the new version number, and successful writes of the new elements are acknowledged by the replicas based at least in part on the new version number.

Abstract: Methods, systems, and computer-readable media for multi-tenant partitioning in a time-series database are disclosed. A partitioning scheme is determined that maps a plurality of data points to a plurality of partitions based at least in part on table identifiers associated with the data points. The partitions are stored using a plurality of storage resources. After the storage resources are provisioned, an additional table identifier is generated. Based at least in part on the partitioning scheme, one or more additional data points comprising the additional table identifier are mapped to a particular partition of the plurality of partitions. The one or more additional data points are stored in the particular partition using the storage resources.

Abstract: Methods, systems, and computer-readable media for splitting partitions across database clusters in a time-series database are disclosed. A time-series database determines that a heat metric for the first tile has exceeded a threshold. The first tile represents spatial boundaries and temporal boundaries of time-series data, and a lease for the first tile is assigned to a storage node. Based (at least in part on) the heat metric, a temporal split of the first tile is performed to generate an intermediate tile representing the spatial boundaries and a later portion of the temporal boundaries. A spatial split of the intermediate tile is performed to generate second and third tiles representing two portions of the spatial boundaries and the later portion of the temporal boundaries. The storage node stores elements of the time-series data within these new boundaries to the second and third tiles.

Abstract: Methods, systems, and computer-readable media for multi-tenant partitioning in a time-series database are disclosed. A partitioning scheme is determined that maps a plurality of data points to a plurality of partitions based at least in part on table identifiers associated with the data points. The partitions are stored using a plurality of storage resources. After the storage resources are provisioned, an additional table identifier is generated. Based at least in part on the partitioning scheme, one or more additional data points comprising the additional table identifier are mapped to a particular partition of the plurality of partitions. The one or more additional data points are stored in the particular partition using the storage resources.

Abstract: Techniques for -memory ingestion for highly available distributed time-series databases are described. A method of in-memory ingestion may include obtaining, by a host of a time series database, time series data from one or more electronic devices, the time series database including a plurality of portions of the time series database spread across a plurality of hosts, the plurality of portions of the time series database including at least one hot portion and a plurality of cold portions, storing the time series data in a volatile storage location associated with the hot portion on the host, detecting an event to store the time series data associated with the hot portion to a non-volatile storage location on the host, and storing the time series data associated with the hot portion to the non-volatile storage location.

Abstract: Methods, systems, and computer-readable media for dynamic lease assignments in a time-series database are disclosed. A time-series database determines an assignment of a lease for a tile representing spatial and temporal boundaries of time-series data. The lease is assigned to a first storage node of a plurality of storage nodes. The time-series database routes the elements of the time-series data within the spatial and temporal boundaries to the first storage node based at least in part on the assignment of the lease. The first storage node stores the elements of the time-series data into the tile in a database cluster. Write requests by the first storage node to the tile are validated by the database cluster based at least in part on the assignment of the lease.

Abstract: Methods, systems, and computer-readable media for ingestion partition auto-scaling in a time-series database are disclosed. A first set of one or more hosts divides elements of time-series data into a plurality of partitions. A second set of one or more hosts stores the elements of time-series data from the plurality of partitions into one or more storage tiers of a time-series database. An analyzer receives first data indicative of the resource usage of the time-series data at the first set of one or more hosts. The analyzer receives second data indicative of the resource usage of the time-series data at the second set of one or more hosts. Based at least in part on analysis of the first data and the second data, the analyzer initiates a split of an individual one of the partitions into two or more partitions.

Abstract: Methods, systems, and computer-readable media for schema-based spatial partitioning in a time-series database are disclosed. A time-series database divides elements of time-series data of a plurality of time series into partitions according to a clustering scheme. The time series are associated with respective schemas. The schemas comprise dimension names and measure names. The clustering scheme is determined based (at least in part) on the schemas and dimension values, and the schemas are determined based (at least in part) on the elements of time-series data. The time-series database stores the elements of time-series data from the plurality of partitions into one or more storage tiers. The time-series database performs a query of the time-series data in one or more of the storage tiers. The query is performed based (at least in part) on the clustering scheme.

Abstract: Creating replicas of a time series database from across storage groups may be implemented for a time series database. Updates to a time series database may be maintained in an update log. Updates may be obtained from the log and ingested at different groups of copies of the time series database used to perform queries. Updates may be ingested at different rates at the different groups. A new copy may be added to one of the groups by copying a portion of the time series database for the new copy determined to be present in another group of copies and an update not found in the other from the log to the new copy.

Abstract: Methods, systems, and computer-readable media for heat balancing in a distributed time-series database are disclosed. A time-series database stores time-series data using database clusters. A plurality of leases for tiles representing spatial and temporal partitions of the time-series data are assigned to a first storage node. The time-series database determines that a heat metric for the first storage node has exceeded a threshold. The time-series database determines respective heat metrics for additional storage nodes including a second storage node. The time-series database selects the second storage node based (at least in part) on the respective heat metrics. The time-series database reassigns one or more of the leases from the first storage node to the second storage node. The second storage node stores elements of the time-series data into the plurality of database clusters in one or more tiles associated with the one or more reassigned leases.

Abstract: Methods, systems, and computer-readable media for partition splitting in a distributed database are disclosed. A partition of data is split into a first sub-partition and a second sub-partition. A first portion of the data is assigned to the first sub-partition, and a second portion of the data is assigned to the second sub-partition. One or more elements of the first portion of the data from the partition and an additional one or more elements of the first portion of the data from the first sub-partition are stored into a first node. One or more elements of the second portion of the data from the partition and an additional one or more elements of the second portion of the data from the second sub-partition are stored into a second node. The partition is prevented from receiving new data after the partition is split into the first and second sub-partitions.

Abstract: Methods, systems, and computer-readable media for ingestion partition auto-scaling in a time-series database are disclosed. A first set of one or more hosts divides elements of time-series data into a plurality of partitions. A second set of one or more hosts stores the elements of time-series data from the plurality of partitions into one or more storage tiers of a time-series database. An analyzer receives first data indicative of the resource usage of the time-series data at the first set of one or more hosts. The analyzer receives second data indicative of the resource usage of the time-series data at the second set of one or more hosts. Based at least in part on analysis of the first data and the second data, the analyzer initiates a split of an individual one of the partitions into two or more partitions.

Abstract: Methods, systems, and computer-readable media for eventually consistent replication in a time-series database are disclosed. Time-series data is stored in a plurality of replicas of a replica group. The time-series data is within spatial and temporal boundaries of the replica group. The replicas store a server-side version number of replica group metadata. A query processor sends a query request to a first replica of the replica group. The query request comprises a query-client-side version number of the replica group metadata. The first replica performs a comparison of the query-client-side version number to the server-side version number. The first replica generates a successful response to the query request or denies the query request based at least in part on the comparison, where the successful response comprises one or more elements of the time-series data.

Abstract: Methods, systems, and computer-readable media for schema-based spatial partitioning in a time-series database are disclosed. A time-series database divides elements of time-series data of a plurality of time series into partitions according to a clustering scheme. The time series are associated with respective schemas. The schemas comprise dimension names and measure names. The clustering scheme is determined based (at least in part) on the schemas and dimension values, and the schemas are determined based (at least in part) on the elements of time-series data. The time-series database stores the elements of time-series data from the plurality of partitions into one or more storage tiers. The time-series database performs a query of the time-series data in one or more of the storage tiers. The query is performed based (at least in part) on the clustering scheme.

Abstract: Transactional log data for a distributed database may be archived by a number of computing nodes scalable to a specified concurrency parameter. A log record may be mapped to a computing node for performance of an archiving operation. The log record may be stored with information indicative of a position, in a hierarchy of shards, of the shard from which the log record originated. The log record may be replayed in an order, relative to other records, that is based on the shard's position in the shard hierarchy. The log record may be replayed on a shard identified based on a time-invariant and idempotent hash function.

Abstract: Methods, systems, and computer-readable media for partition segmenting in a distributed time-series database are disclosed. A first host receives a request to store a data element in a tile comprising linked segments including a head segment and a tail segment. The head and tail segments are associated with the same spatial and temporal boundaries, and the data element is within those boundaries. The first host determines that the data element is not stored in the head segment and forwards the data element to a next segment of the tile. An additional host receives a request to store the data element in the tail segment. The additional host determines that the data element is not stored in the tail segment and then stores the data element in the tail segment.