Granularity and Specifications
DataJob Granularity
BladePipe DataJob refer to configuration groups that complete a data migration, synchronization, verification, correction, and other goals for a specific database, and perform actual actions successively through a set of specific processes.
BladePipe DataJob is also called Link.
Individual DataJob migrate or synchronize maximum and minimum database scopes (granularity) depending on the database type. This is shown in the following table.
Source DataSource | The Maximum Granularity of a Link | The Minimum Granularity of a Link | Active Rows Limited |
---|---|---|---|
MySQL | Multiple Schemas | Single Table | Unlimited |
Oracle | Single Schema | Single Table | Unlimited |
SQLServer | Single Schema | Single Table | Unlimited |
Db2 | Single Schema | Single Table | Unlimited |
PostgreSQL | Multiple Schemas | Single Table | Unlimited |
MongoDB | Multiple Schemas | Single Collection | Unlimited |
OceanBase | Multiple Schemas | Single Table | Unlimited |
PolarDbX | Multiple Schemas | Single Table | Unlimited |
StarRocks | Multiple Schemas | Single Table | Unlimited |
Tunnel | Multiple Schemas | Single Table | Unlimited |
TiDB | Multiple Schemas | Single Table | Unlimited |
Kafka | Multiple Topics | Single Topic | Unlimited |
RocketMQ | Multiple Topics | Single Topic | Unlimited |
RabbitMQ | Multiple Queues | Single Queue | Unlimited |
AutoMQ | Multiple Topics | Single Topic | Unlimited |
Hana | Single Schema | Single Table | Unlimited |
OpenGauss | Multiple Schemas | Single Table | Unlimited |
Greenplum | Multiple Schemas | Single Table | Unlimited |
Elasticsearch | Multiple Indexes | Single Index | Unlimited |
Pulsar | Multiple Topics | Single Topic | Unlimited |
TDengine | Multiple Schemas | Single Table | Unlimited |
Dameng | Multiple Schemas | Single Table | Unlimited |
SshFile | Single Catalog | Single File | Unlimited |
S3File | Single Bucket | Single File | Unlimited |
OssFile | Single Bucket | Single File | Unlimited |
DataJob Specifications
When a BladePipe DataJob uses a specific set of processes to sequentially complete a data migration / synchronization work, the combination of memory footprint when these processes run is called a specification.
There are four types of processes: Full Data, Incremental, Verification, and Correction. The following table briefly describes the specifications that comprise these process memory configurations.
Specification Number | Series | Process Memory Size |
---|---|---|
1 | Incremental Enhanced | Full Data 1GB,Incremental 2GB,Verification 1GB,Correction 1GB |
2 | Incremental Enhanced | Full Data 1GB,Incremental 3GB,Verification 1GB,Correction 1GB |
3 | Incremental Enhanced | Full Data 1GB,Incremental 4GB,Verification 1GB,Correction 1GB |
4 | Incremental Enhanced | Full Data 2GB,Incremental 3GB,Verification 2GB,Correction 2GB |
5 | Incremental Enhanced | Full Data 2GB,Incremental 4GB,Verification 2GB,Correction 2GB |
6 | Full Data Enhanced | Full Data 1GB,Incremental 0.5GB,Verification 1GB,Correction 1GB |
7 | Full Data Enhanced | Full Data 2GB,Incremental 0.5GB,Verification 2GB,Correction 2GB |
8 | Full Data Enhanced | Full Data 3GB,Incremental 0.5GB,Verification 3GB,Correction 3GB |
9 | Full Data Enhanced | Full Data 4GB,Incremental 0.5GB,Verification 4GB,Correction 4GB |
10 | Full Data Enhanced | Full Data 2GB,Incremental 1GB,Verification 2GB,Correction 2GB |
11 | Full Data Enhanced | Full Data 3GB,Incremental 1GB,Verification 3GB,Correction 3GB |
12 | Full Data Enhanced | Full Data 4GB,Incremental 1GB,Verification 4GB,Correction 4GB |
13 | Full Data Enhanced | Full Data 3GB,Incremental 2GB,Verification 3GB,Correction 3GB |
14 | Full Data Enhanced | Full Data 4GB,Incremental 2GB,Verification 4GB,Correction 4GB |
15 | Balanced | Full Data 1GB,Incremental 1GB,Verification 1GB,Correction 1GB |
16 | Balanced | Full Data 2GB,Incremental 2GB,Verification 2GB,Correction 2GB |
17 | Balanced | Full Data 3GB,Incremental 3GB,Verification 3GB,Correction 3GB |
18 | Balanced | Full Data 4GB,Incremental 4GB,Verification 4GB,Correction 4GB |
19 | Balanced | Full Data 0.5GB,Incremental 0.5GB,Verification 0.5GB,Correction 0.5GB |
20 | Incremental Enhanced | Full Data 2GB,Incremental 5GB,Verification 2GB,Correction 2GB |
21 | Incremental Enhanced | Full Data 2GB,Incremental 6GB,Verification 2GB,Correction 2GB |
22 | Incremental Enhanced | Full Data 2GB,Incremental 7GB,Verification 2GB,Correction 2GB |
23 | Incremental Enhanced | Full Data 2GB,Incremental 8GB,Verification 2GB,Correction 2GB |
24 | Incremental Enhanced | Full Data 2GB,Incremental 12GB,Verification 2GB,Correction 2GB |
25 | Incremental Enhanced | Full Data 2GB,Incremental 16GB,Verification 2GB,Correction 2GB |
26 | Incremental Enhanced | Full Data 2GB,Incremental 20GB,Verification 2GB,Correction 2GB |
27 | Full Data Enhanced | Full Data 8GB,Incremental 2GB,Verification 2GB,Correction 2GB |
28 | Full Data Enhanced | Full Data 16GB,Incremental 2GB,Verification 2GB,Correction 2GB |
Note:
- A balanced version of specification number 16 is recommended and will meet most requirements.
- The 0.5GB memory associated specification is not recommended for production use.
- Sub-DataJob generated by editing subscriptions automatically inherits the specifications of the parent DataJob.
- Only specification numbers 15/16/17/18/19/20/21/22/23/24/25/26 can be used by Modify Parameters.
- Incremental DataTask pre-occupy machine memory (logical), and the occupied memory (logical) is released after a single Full Data/Verification DataTask is completed.