Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Replace node number with node count #20549

Open
wants to merge 1 commit into
base: master
Choose a base branch
from
Open
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
30 changes: 15 additions & 15 deletions tidb-cloud/size-your-cluster.md
Original file line number Diff line number Diff line change
Expand Up @@ -15,7 +15,7 @@ This document describes how to determine the size of a TiDB Cloud Dedicated clus

TiDB is for computing only and does not store data. It is horizontally scalable.

You can configure node number, vCPU, and RAM for TiDB.
You can configure node count, vCPU, and RAM for TiDB.

To learn performance test results of different cluster scales, see [TiDB Cloud Performance Reference](/tidb-cloud/tidb-cloud-performance-reference.md).

Expand All @@ -36,11 +36,11 @@ The supported vCPU and RAM sizes include the following:
>
> If the vCPU and RAM size of TiDB is set as **4 vCPU, 16 GiB**, note the following restrictions:
>
> - The node number of TiDB can only be set to 1 or 2, and the node number of TiKV is fixed to 3.
> - The node count of TiDB can only be set to 1 or 2, and the node count of TiKV is fixed to 3.
> - 4 vCPU TiDB can only be used with 4 vCPU TiKV.
> - TiFlash is unavailable.

### TiDB node number
### TiDB node count

For high availability, it is recommended that you configure at least two TiDB nodes for each TiDB Cloud cluster.

Expand All @@ -67,9 +67,9 @@ When planning your cluster size, you can estimate the number of TiDB nodes accor

`node num = ceil(overall expected performance ÷ performance per node * (1 - performance deviation coefficient))`

In the formula, you need to calculate `node num = ceil(overall expected performance ÷ performance per node)` first to get a rough node number, and then use the corresponding performance deviation coefficient to get the final result of the node number.
In the formula, you need to calculate `node num = ceil(overall expected performance ÷ performance per node)` first to get a rough node count, and then use the corresponding performance deviation coefficient to get the final result of the node count.

For example, your overall expected performance is 110,000 QPS under a mixed workload, your P95 latency is about 100 ms, and you want to use 8 vCPU, 16 GiB TiDB nodes. Then, you can get the estimated TiDB performance of an 8 vCPU, 16 GiB TiDB node from the preceding table (which is `15,500`), and calculate a rough number of TiDB nodes as follows:
For example, your overall expected performance is 110,000 QPS under a mixed workload, your P95 latency is about 100 ms, and you want to use 8 vCPU, 16 GiB TiDB nodes. Then, you can get the estimated TiDB performance of an 8 vCPU, 16 GiB TiDB node from the preceding table (which is `15,500`), and calculate a rough count of TiDB nodes as follows:

`node num = ceil(110,000 ÷ 15,500) = 8`

Expand All @@ -81,7 +81,7 @@ Therefore, 8 TiDB nodes (8 vCPU, 16 GiB) are recommended for you.

TiKV is responsible for storing data. It is horizontally scalable.

You can configure node number, vCPU and RAM, and storage for TiKV.
You can configure node count, vCPU and RAM, and storage for TiKV.

To learn performance test results of different cluster scales, see [TiDB Cloud Performance Reference](/tidb-cloud/tidb-cloud-performance-reference.md).

Expand All @@ -100,11 +100,11 @@ The supported vCPU and RAM sizes include the following:
>
> If the vCPU and RAM size of TiKV is set as **4 vCPU, 16 GiB**, note the following restrictions:
>
> - The node number of TiDB can only be set to 1 or 2, and the node number of TiKV is fixed to 3.
> - The node count of TiDB can only be set to 1 or 2, and the node count of TiKV is fixed to 3.
> - 4 vCPU TiKV can only be used with 4 vCPU TiDB.
> - TiFlash is unavailable.

### TiKV node number
### TiKV node count

The number of TiKV nodes should be **at least 1 set (3 nodes in 3 different Available Zones)**.

Expand All @@ -114,9 +114,9 @@ TiDB Cloud deploys TiKV nodes evenly to all availability zones (at least 3) in t
>
> When you scale your TiDB cluster, nodes in the 3 availability zones are increased or decreased at the same time. For how to scale in or scale out a TiDB cluster based on your needs, see [Scale Your TiDB Cluster](/tidb-cloud/scale-tidb-cluster.md).

Although TiKV is mainly used for data storage, the performance of the TiKV node also varies depending on different workloads. Therefore, when planning the number of TiKV nodes, you need to estimate it according to both your [**data volume**](#estimate-tikv-node-number-according-to-data-volume) and [expected performance](#estimate-tikv-node-number-according-to-expected-performance), and then take the larger of the two estimates as the recommended node number.
Although TiKV is mainly used for data storage, the performance of the TiKV node also varies depending on different workloads. Therefore, when planning the number of TiKV nodes, you need to estimate it according to both your [**data volume**](#estimate-tikv-node-number-according-to-data-volume) and [expected performance](#estimate-tikv-node-number-according-to-expected-performance), and then take the larger of the two estimates as the recommended node count.

#### Estimate TiKV node number according to data volume
#### Estimate TiKV node nucountmber according to data volume

You can calculate a recommended number of TiKV nodes according to your data volume as follows:

Expand All @@ -130,7 +130,7 @@ Suppose that the size of your MySQL dump files is 20 TB and the TiKV compression

`node num = ceil(20 TB * 40% * 3 ÷ 0.8 ÷ 4096 GiB ÷ 3) * 3 = 9`

#### Estimate TiKV node number according to expected performance
#### Estimate TiKV node count according to expected performance

Similarly as TiDB performance, TiKV performance increases linearly with the number of TiKV nodes. However, when the number of TiKV nodes exceeds 8, the performance increment becomes slightly less than linearly proportional. For each additional 8 nodes, the performance deviation coefficient is about 5%.

Expand All @@ -155,7 +155,7 @@ When planning your cluster size, you can estimate the number of TiKV nodes accor

`node num = ceil(overall expected performance ÷ performance per node * (1 - performance deviation coefficient))`

In the formula, you need to calculate `node num = ceil(overall expected performance ÷ performance per node)` first to get a rough node number, and then use the corresponding performance deviation coefficient to get the final result of the node number.
In the formula, you need to calculate `node num = ceil(overall expected performance ÷ performance per node)` first to get a rough node number, and then use the corresponding performance deviation coefficient to get the final result of the node count.

For example, your overall expected performance is 110,000 QPS under a mixed workload, your P95 latency is about 100 ms, and you want to use 8 vCPU, 32 GiB TiKV nodes. Then, you can get the estimated TiKV performance of an 8 vCPU, 32 GiB TiKV node from the preceding table (which is `17,800`), and calculate a rough number of TiKV nodes as follows:

Expand All @@ -165,7 +165,7 @@ As 7 is less than 8, the performance deviation coefficient of 7 nodes is 0. The

Therefore, 7 TiKV nodes (8 vCPU, 32 GiB) are recommended for you according to your expected performance.

Next, you can compare the TiKV node number calculated according to data volume with the number calculated according to your expected performance, and take the larger one as a recommended number of your TiKV nodes.
Next, you can compare the TiKV node count calculated according to data volume with the number calculated according to your expected performance, and take the larger one as a recommended number of your TiKV nodes.

### TiKV node storage

Expand All @@ -186,7 +186,7 @@ The supported node storage of different TiKV vCPUs is as follows:

TiFlash synchronizes data from TiKV in real time and supports real-time analytics workloads right out of the box. It is horizontally scalable.

You can configure node number, vCPU and RAM, and storage for TiFlash.
You can configure node count, vCPU and RAM, and storage for TiFlash.

### TiFlash vCPU and RAM

Expand All @@ -198,7 +198,7 @@ The supported vCPU and RAM sizes include the following:

Note that TiFlash is unavailable when the vCPU and RAM size of TiDB or TiKV is set as **4 vCPU, 16 GiB**.

### TiFlash node number
### TiFlash node count

TiDB Cloud deploys TiFlash nodes evenly to different availability zones in a region. It is recommended that you configure at least two TiFlash nodes in each TiDB Cloud cluster and create at least two replicas of the data for high availability in your production environment.

Expand Down
Loading