tencent cloud

All product documents
TDMQ for Apache Pulsar
DocumentationTDMQ for Apache PulsarPurchase GuidePro ClusterFixed Storage
Fixed Storage
Download PDF
Last updated: 2025-04-01 11:16:45
Fixed Storage
Last updated: 2025-04-01 11:16:45
Download PDF
This document mainly describes the specifications of the TDMQ for Apache Pulsar Pro cluster (fixed storage), including TPS, bandwidth, and topic quantity limit. They are used for your selection and usage.
Note:
1. Pro cluster storage uses a triple replica mechanism to ensure data high availability.
2. Pro clusters support online upgrade configuration. For specific operations, see the document Upgrading Cluster. Theoretically, the upgrade process should be seamless for the business. However, during server-side Rebalance Topic operations, clients may experience disconnections followed by reconnections.

Pro Cluster Specification

No.
Cluster Specification Code
Max Messaging TPS
Peak Bandwidth (MB/s)
Max Topics per Cluster
Min Storage (GB)
1
PULSAR.P1.MINI2
2000
45
1000
200
2
PULSAR.P1.SMALL4
4000
90
1000
200
3
PULSAR.P1.SMALL6
6000
120
1000
267
4
PULSAR.P1.SMALL10
10000
180
2000
400
5
PULSAR.P1.MEDIUM15
15000
300
2000
333
6
PULSAR.P1.MEDIUM20
20000
480
2000
400
7
PULSAR.P1.MEDIUM40
40000
720
2000
533
8
PULSAR.P1.MEDIUM60
60000
1080
2000
533
9
PULSAR.P1.MEDIUM100
100000
1920
3000
533
10
PULSAR.P1.LARGE150
150000
2400
3000
667
11
PULSAR.P1.LARGE200
200000
3600
3000
800
12
PULSAR.P1.LARGE400
400000
4200
3000
1000
13
PULSAR.P1.LARGE600
600000
4800
3000
1200
14
PULSAR.P1.LARGE1000
1000000
6000
3000
1667

Notes

The max messaging TPS is calculated based on the general message type and message size of 4 KB. The values for advanced messages and large messages need to be multiplied accordingly. The peak bandwidth indicates the maximum inbound and outbound bandwidth. For specific calculation methods, see Pro Cluster Billing.

Storage Usage Instructions

1. Professional Cluster Fixed Storage Edition. The storage usage statistics are about the usage of Disks in the storage service Bookie, not the storage usage of topic data in Brokers.
2. Cannot calculate the disk usage of the storage service Bookie based on the storage usage of all business topics in the Broker. Because the storage space used by topics only accounts for part of the Bookie storage. There are also index files, log files, etc. in the Bookie storage that will occupy space.
3. Not only business topic data occupies the Bookie disk, but also Cursor, Schema and internal system topics also occupy the Bookie disk.
4. Under the data recovery mechanism of Bookie, when the data of topics in the Broker is deleted, the disk in Bookie may not be released immediately. Write amplification may occur here. Usually, write amplification is within 30%, but in extreme cases, it may reach 50%. This has something to do with the data clearing (called GC internally in Bookie) mechanism of Bookie.
5. It is recommended that you configure and pay close attention to cluster storage utilization alarms. It is advisable to control the storage utilization within 80% in production environments to prevent partial nodes from being filled up in advance due to unbalanced node storage usage.
Was this page helpful?
You can also Contact Sales or Submit a Ticket for help.
Yes
No

Feedback

Contact Us

Contact our sales team or business advisors to help your business.

Contact Us
Technical Support

Open a ticket if you're looking for further assistance. Our Ticket is 7x24 available.

Submit a Ticket
7x24 Phone Support
Hong Kong, China
+852 800 906 020 (Toll Free)
United States
+1 844 606 0804 (Toll Free)
United Kingdom
+44 808 196 4551 (Toll Free)
Canada
+1 888 605 7930 (Toll Free)
Australia
+61 1300 986 386 (Toll Free)
EdgeOne hotline
+852 300 80699
More local hotlines coming soon