LindormDFS is a distributed file storage service for large amounts of unstructured data. LindormDFS shares the underlying storage with multiple engines. Lindorm provides database-level security and reliability. LindormDFS is fully compatible with open source Hadoop Distributed File System (HDFS) communication protocols. You can use open source HDFS clients to directly access LindormDFS. This way, LindormDFS can seamlessly connect to open source ecosystems of big data and cloud computing.
Select storage specifications
If you use LindormDFS as the underlying storage for big data and data lakes, we recommend that you select Capacity for the Storage Type parameter.
You are billed for the storage of ApsaraDB for Lindorm (Lindorm) based on the reserved storage. Make sure that the reserved storage of LindormDFS exceeds your maximum used storage by more than 10%.
The cost of the capacity-optimized storage type is low. The price in each region prevails. In addition, this storage type includes the sequential read and write capabilities that provide high throughput. However, the random read capability of the capacity-optimized storage type is weak. The capacity-optimized storage type is suitable for big data computing scenarios.
Select node specifications
LindormDFS nodes are responsible for the read and write bandwidth. The nodes are used to manage data blocks and provide support for HDFS communication protocols. You can determine the number of LindormDFS nodes in your business scenarios from the following two dimensions:
Data volume: Each LindormDFS node supports a storage capacity in the range of 10 TB to 50 TB.
Bandwidth: Each LindormDFS node supports a data transmission bandwidth in the range of 100 MB/s to 200 MB/s.
You can calculate the required number of LindormDFS nodes from the data volume dimension and bandwidth dimension. You can use the upper limit of the calculation to identify the minimum number of nodes that you need to purchase.