You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Copy file name to clipboardExpand all lines: src/UserGuide/Master/Table/Deployment-and-Maintenance/Database-Resources.md
+59-43Lines changed: 59 additions & 43 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -19,50 +19,50 @@
19
19
20
20
-->
21
21
# Database Resources
22
-
## CPU
22
+
## 1 CPU
23
23
<tablestyle="text-align: center;">
24
24
<tbody>
25
25
<tr>
26
-
<th rowspan="2">Number of timeseries (frequency<=1HZ)</th>
26
+
<th rowspan="2">Number of timeseries (frequency<=1HZ)</th>
27
27
<th rowspan="2">CPU</th>
28
28
<th colspan="3">Number of nodes</th>
29
29
</tr>
30
30
<tr>
31
-
<th>standalone mode</th>
32
-
<th>Double active</th>
31
+
<th>standalone</th>
32
+
<th>Dual-Active</th>
33
33
<th>Distributed</th>
34
34
</tr>
35
35
<tr>
36
36
<td>Within 100000</td>
37
-
<td>2core-4core</td>
37
+
<td>2-4 cores</td>
38
38
<td>1</td>
39
39
<td>2</td>
40
40
<td>3</td>
41
41
</tr>
42
42
<tr>
43
43
<td>Within 300000</td>
44
-
<td>4core-8core</td>
44
+
<td>4-8 cores</td>
45
45
<td>1</td>
46
46
<td>2</td>
47
47
<td>3</td>
48
48
</tr>
49
49
<tr>
50
50
<td>Within 500000</td>
51
-
<td>8core-26core</td>
51
+
<td>8-16 cores</td>
52
52
<td>1</td>
53
53
<td>2</td>
54
54
<td>3</td>
55
55
</tr>
56
56
<tr>
57
57
<td>Within 1000000</td>
58
-
<td>16core-32core</td>
58
+
<td>16-32 cores</td>
59
59
<td>1</td>
60
60
<td>2</td>
61
61
<td>3</td>
62
62
</tr>
63
63
<tr>
64
64
<td>Within 2000000</td>
65
-
<td>32core-48core</td>
65
+
<td>32-48 cores</td>
66
66
<td>1</td>
67
67
<td>2</td>
68
68
<td>3</td>
@@ -81,50 +81,50 @@
81
81
</tbody>
82
82
</table>
83
83
84
-
## Memory
84
+
## 2 Memory
85
85
<tablestyle="text-align: center;">
86
86
<tbody>
87
87
<tr>
88
-
<th rowspan="2">Number of timeseries (frequency<=1HZ)</th>
88
+
<th rowspan="2">Number of timeseries (frequency<=1HZ)</th>
89
89
<th rowspan="2">Memory</th>
90
90
<th colspan="3">Number of nodes</th>
91
91
</tr>
92
92
<tr>
93
-
<th>standalone mode</th>
94
-
<th>Double active</th>
93
+
<th>standalone</th>
94
+
<th>Dual-Active</th>
95
95
<th>Distributed</th>
96
96
</tr>
97
97
<tr>
98
98
<td>Within 100000</td>
99
-
<td>4G-8G</td>
99
+
<td>4-8G</td>
100
100
<td>1</td>
101
101
<td>2</td>
102
102
<td>3</td>
103
103
</tr>
104
104
<tr>
105
105
<td>Within 300000</td>
106
-
<td>12G-32G</td>
106
+
<td>12-32G</td>
107
107
<td>1</td>
108
108
<td>2</td>
109
109
<td>3</td>
110
110
</tr>
111
111
<tr>
112
112
<td>Within 500000</td>
113
-
<td>24G-48G</td>
113
+
<td>24-48G</td>
114
114
<td>1</td>
115
115
<td>2</td>
116
116
<td>3</td>
117
117
</tr>
118
118
<tr>
119
119
<td>Within 1000000</td>
120
-
<td>32G-96G</td>
120
+
<td>32-96G</td>
121
121
<td>1</td>
122
122
<td>2</td>
123
123
<td>3</td>
124
124
</tr>
125
125
<tr>
126
126
<td>Within 2000000</td>
127
-
<td>64G-128G</td>
127
+
<td>64-128G</td>
128
128
<td>1</td>
129
129
<td>2</td>
130
130
<td>3</td>
@@ -143,19 +143,23 @@
143
143
</tbody>
144
144
</table>
145
145
146
-
## Storage (Disk)
147
-
### Storage space
148
-
Calculation formula: Number of measurement points * Sampling frequency (Hz) * Size of each data point (Byte, different data types may vary, see table below) * Storage time (seconds) * Number of copies (usually 1 copy for a single node and 2 copies for a cluster) ÷ Compression ratio (can be estimated at 5-10 times, but may be higher in actual situations)
146
+
## 3 Storage (Disk)
147
+
### 3.1 Storage space
148
+
Calculation Formula:
149
+
150
+
```Plain
151
+
Storage Space = Number of Measurement Points * Sampling Frequency (Hz) * Size of Each Data Point (Bytes, see the table below) * Storage Duration * Replication Factor / Compression Ratio
152
+
```
153
+
154
+
Data Point Size Calculation Table:
155
+
149
156
<tablestyle="text-align: center;">
150
157
<tbody>
151
158
<tr>
152
-
<th colspan="4">Data point size calculation</th>
153
-
</tr>
154
-
<tr>
155
-
<th>data type</th>
159
+
<th>Data Type</th>
156
160
<th>Timestamp (Bytes)</th>
157
161
<th> Value (Bytes)</th>
158
-
<th> Total size of data points (in bytes)
162
+
<th> Total Data Point Size (Bytes)
159
163
</th>
160
164
</tr>
161
165
<tr>
@@ -165,36 +169,48 @@ Calculation formula: Number of measurement points * Sampling frequency (Hz) * Si
165
169
<td>9</td>
166
170
</tr>
167
171
<tr>
168
-
<td> INT32/FLOAT</td>
172
+
<td> INT32 / FLOAT (Single Precision)</td>
169
173
<td>8</td>
170
174
<td>4</td>
171
175
<td>12</td>
172
176
</tr>
173
177
<tr>
174
-
<td>INT64/DOUBLE</td>
178
+
<td>INT64 / DOUBLE (Double Precision)</td>
175
179
<td>8</td>
176
180
<td>8</td>
177
181
<td>16</td>
178
182
</tr>
179
183
<tr>
180
-
<td>TEXT</td>
184
+
<td>TEXT (String)</td>
181
185
<td>8</td>
182
-
<td>The average is a</td>
186
+
<td>Average = a</td>
183
187
<td>8+a</td>
184
188
</tr>
185
189
</tbody>
186
190
</table>
191
+
Example:
192
+
193
+
- Scenario: 1,000 devices, 100 measurement points per device, i.e. 100,000 sequences in total. Data type is INT32. Sampling frequency is 1Hz (once per second). Storage duration is 1 year. Replication factor is 3.
194
+
- Full Calculation:
195
+
```Plain
196
+
1,000 devices * 100 measurement points * 12 bytes per data point * 86,400 seconds per day * 365 days per year * 3 replicas / 10 compression ratio = 11 TB
197
+
```
198
+
- Simplified Calculation:
199
+
```Plain
200
+
1,000 * 100 * 12 * 86,400 * 365 * 3 / 10 = 11 TB
201
+
```
202
+
### 3.2 Storage Configuration
203
+
204
+
- For systems with > 10 million measurement points or high query loads, SSD is recommended.
205
+
206
+
## 4 Network (NIC)
207
+
When the write throughput does not exceed 10 million points per second, a gigabit network card is required. When the write throughput exceeds 10 million points per second, a 10-gigabit network card is required.
Example: 1000 devices, each with 100 measurement points, a total of 100000 sequences, INT32 type. Sampling frequency 1Hz (once per second), storage for 1 year, 3 copies.
189
-
- Complete calculation formula: 1000 devices * 100 measurement points * 12 bytes per data point * 86400 seconds per day * 365 days per year * 3 copies / 10 compression ratio / 1024 / 1024 / 1024 / 1024 =11T
If the number of nodes is over 10000000 or the query load is high, it is recommended to configure SSD
193
-
## Network (Network card)
194
-
If the write throughput does not exceed 10 million points/second, configure 1Gbps network card. When the write throughput exceeds 10 million points per second, a 10Gbps network card needs to be configured.
195
-
|**Write throughput (data points per second)**|**NIC rate**|
196
-
| ------------------- | ------------- |
197
-
| <10 million | 1Gbps |
198
-
| >=10 million | 10Gbps |
199
-
## Other instructions
200
-
IoTDB has the ability to scale up clusters in seconds, and expanding node data does not require migration. Therefore, you do not need to worry about the limited cluster capacity estimated based on existing data. In the future, you can add new nodes to the cluster when you need to scale up.
216
+
- IoTDB supports second-level cluster scaling . Data migration is not required when adding new nodes, so there is no need to worry about limited cluster capacity based on current data estimates. You can add new nodes to the cluster when scaling is needed in the future.
An IoTDB cluster instance consists of 3 ConfigNodes and no fewer than 3 DataNodes, typically 3 DataNodes, referred to as 3C3D. In the event of partial node failures, the remaining nodes can still provide services, ensuring high availability of the database service, and the database performance can be improved with the addition of nodes.
37
+
38
+
-**Features**: High availability and scalability, with the ability to enhance system performance by adding DataNodes.
39
+
-**Applicable scenarios**: Enterprise-level application scenarios that require high availability and reliability.
| Applicable Scenario | Edge deployment, low requirement for high availability | High-availability business, disaster recovery scenarios, etc. |
47
+
| Number of Machines Required | 1 | ≥3 |
48
+
| Security and Reliability | Cannot tolerate single-point failure | High, can tolerate single-point failure |
49
+
| Scalability | Scalable by adding DataNodes to improve performance | Scalable by adding DataNodes to improve performance |
50
+
| Performance | Scalable with the number of DataNodes | Scalable with the number of DataNodes |
51
+
52
+
- The deployment steps for standalone mode and cluster mode are similar (adding ConfigNodes and DataNodes one by one), with the only differences being the number of replicas and the minimum number of nodes required to provide services.
Dual-Active Deployment is a feature of TimechoDB, where two independent instances synchronize bidirectionally and can provide services simultaneously. If one instance stops and restarts, the other instance will resume data transfer from the breakpoint.
36
+
37
+
> An IoTDB Dual-Active instance typically consists of 2 standalone nodes, i.e., 2 sets of 1C1D. Each instance can also be a cluster.
38
+
39
+
-**Features**: The high-availability solution with the lowest resource consumption.
40
+
-**Use Cases**: Scenarios with limited resources (only two servers) but requiring high availability.
An IoTDB cluster instance consists of 3 ConfigNodes and no fewer than 3 DataNodes, typically 3 DataNodes, i.e., 3C3D. If some nodes fail, the remaining nodes can still provide services, ensuring high availability of the database. Performance can be improved by adding DataNodes.
46
+
47
+
-**Features**: High availability, high scalability, and improved system performance by adding DataNodes.
48
+
-**Use Cases**: Enterprise-level application scenarios requiring high availability and reliability.
| Security and Reliability | Cannot tolerate single-point failure | High, can tolerate single-point failure | High, can tolerate single-point failure |
60
+
| Scalability | Can expand DataNodes to improve performance | Each instance can be scaled as needed | Can expand DataNodes to improve performance |
61
+
| Performance | Can scale with the number of DataNodes | Same as one of the instances | Can scale with the number of DataNodes |
62
+
63
+
- The deployment steps for Stand-Alone Mode and Cluster Mode are similar (adding ConfigNodes and DataNodes one by one), with differences only in the number of replicas and the minimum number of nodes required to provide services.
0 commit comments