Distributed time sequence database system, electronic equipment and storage medium

Abstract

The invention belongs to the technical field of distributed databases, and particularly discloses a distributed time sequence database system, electronic equipment and a storage medium. The system comprises a metadata service, an intermediate layer service, an intelligent client, a storage layer service and a management service, wherein the metadata service is used for storing metadata of each assembly of the database system; the intermediate layer service is used for determining a routing strategy of the data and determining physical addresses distributed by a plurality of data copies; the intelligent client is used for initiating a data read-write request, comprises a load balancing algorithm and is used for selecting the intermediate layer service; the storage layer service is used for storing a plurality of data copies; and a management service (6) which manages the metadata used in the metadata service. The system has the advantages that 1, the system does not depend on an additional network environment, can be deployed in any intercommunication network, and supports cloud native deployment; and 2, each level can be expanded infinitely, capacity bottlenecks and performance bottlenecks of single nodes do not exist, and the overall performance of the cluster can be improved through the expandability.

Description

technical field [0001] The invention relates to the technical field of distributed databases, in particular to a distributed time series database system, electronic equipment and storage media. Background technique [0002] At present, many companies in the industry use Influxdb as a time-series database internally. Generally, they use the double-write mode based on the Influxdb-Proxy service or the third-party storage mode based on distributed block storage or distributed file storage, or use the above two modes in combination. In the double-write mode based on the Influxdb-Proxy service, after the data is written into the Influxb-Proxy service, it is copied and split into two copies, and double-written to two separate Influxdb nodes. The two nodes store the same data as backups for each other. In the third-party storage mode based on distributed block storage or distributed file storage, the physical disk where Influxdb is stored is replaced with a distributed cloud disk, ...

AI Technical Summary

Problems solved by technology

[0004] 1. To use the double-write mode of the Influxdb-Proxy mode, other services need to be used to further ensure the high availability of the Influxdb-Proxy service. If there is an exception in the single-point InfluxdbProxy, it will cause the entire timing storage layer service to fail

[0005] 2. A single-point Influxdb that does not use cloud disks as storage will reach a physical limit in capacity, and cannot increase the capacity of the entire cluster in a scalable manner;

[0006] 3. To use cloud disks as Influxdb storage, you must rely on third-party distributed block storage or distributed file storage layer services, which are still operable under the premise that cloud vendors provide this service. If you deploy and maintain distributed storage by yourself Layer services will bring huge manpower and operation and maintenance costs

[0007] 4. Neither the double-write mode nor the cloud disk-based mode can solve the problem that may lead to a single Influxdb node read-write performance bottleneck in large-scale data read-write scenarios

If you create additional instances, you need complex control at the business level, and its scalability is quite limited

Images