Data distribution control method, system and device of a distributed storage system
A distributed storage and data distribution technology, which is applied in database distribution/replication, transmission system, electronic digital data processing, etc., can solve the problems of reducing network scale, network burden, and high cost of network equipment and architecture adjustment, and achieve improvement Read and write performance, the effect of increasing bandwidth costs
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0026] figure 1 Shown is the layered network architecture of the storage nodes in the distributed storage system of this embodiment (also referred to as the system herein), including the following nodes:
[0027] A storage node, a storage node is an entity that stores data, and is also called a machine (host, Machine).
[0028] An access switch (ASW) is a switch that provides network access for storage nodes. In actual deployment, multiple machines on one rack are usually connected to the same access switch.
[0029] The aggregation switch (PSW), which is the aggregation point of multiple access layer switches, handles all traffic from the access layer devices and provides uplinks to the core layer.
[0030] Core switch (DSW), the switch deployed at the core layer (network backbone) is called a core switch. The core layer provides an optimized and reliable backbone transmission structure through high-speed forwarding communication.
[0031] The distributed storage system in...
Embodiment 2
[0111] Figure 4 It is the topology diagram of storage nodes in the distributed storage system in this embodiment, including storage nodes, access switches and core switches, please refer to the figure 1 . In this embodiment, the fault-tolerant domain is divided into three layers, ie, Machine, Rack, and Zone. The distributed storage system includes multiple Zones, ZoneA and ZoneB are shown in the figure, and each Zone includes multiple Racks.
[0112] Based on the above topology, each process of the data distribution control method in this embodiment is described below.
[0113] The data distribution control method in this embodiment includes a process of establishing a topology relationship, a process of writing data, and a process of restoring data. in:
[0114] The process of establishing a topology relationship includes:
[0115] Step 1, the storage node generates topology information, and carries the topology information when registering, the topology information incl...
Embodiment 3
[0141] This embodiment relates to the data distribution control of the upper application system. The upper layer application system of this embodiment takes the cloud disk system used by the user as an example to illustrate how to perform distributed control when storing data. The cloud disk system uses the distributed storage system as the underlying storage system, for example, the distributed storage system in Embodiment 1 or Embodiment 2 can be used.
[0142] In the application scenario where users store data through cloud disks, the corresponding data distribution control methods include:
[0143] Step 1. The cloud disk system provides users with various cloud disks with different availability: high-availability cloud disks and low-availability cloud disks;
[0144] High-availability cloud disks can be charged, and users can be promised availability indicators such as failure recovery time. Low-availability cloud disks are free, and no promises are made to users.
[014...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More