mirror of
https://github.com/seaweedfs/seaweedfs.git
synced 2024-01-19 02:48:24 +00:00
Updated FUSE Mount (markdown)
parent
edf6ea798e
commit
0e673e7ba8
|
@ -72,16 +72,17 @@ filer ---- mount1
|
||||||
`weed mount` has a persistent client connecting to Master, to get the location updates of all volumes.
|
`weed mount` has a persistent client connecting to Master, to get the location updates of all volumes.
|
||||||
There are no network round trip to lookup the volume id location.
|
There are no network round trip to lookup the volume id location.
|
||||||
|
|
||||||
|
`weed mount` also subscribe all metadata updates from the Filer. So later reads would not need a network read from Filer, and the metadata reads, e.g., directory listings, are all local operations.
|
||||||
|
|
||||||
For reads:
|
For reads:
|
||||||
1. Client Reads File Metadata => Weed Filer => Weed Filer database (LevelDB, Cassandra, Redis, Mysql, Postgres, etc)
|
1. Mount optionally lookups volume Id => Weed Filer => Weed Master
|
||||||
2. Client optionally lookup volume Id => Weed Filer => Weed Master
|
2. Mount Reads File Chunks => Weed Volume Servers
|
||||||
3. Client Reads File Chunks => Weed Volume Servers
|
|
||||||
|
|
||||||
![](MountRead.png)
|
![](MountRead.png)
|
||||||
|
|
||||||
For writes:
|
For writes:
|
||||||
1. Client uploads data to Weed Volume Servers, and breaks the large files into chunks.
|
1. Mount uploads data to Weed Volume Servers, and breaks the large files into chunks.
|
||||||
2. Client writes the metadata and chunk information into Filer and then into Filer database.
|
2. Mount writes the metadata and chunk information into Filer and then into Filer database.
|
||||||
|
|
||||||
![](MountWrite.png)
|
![](MountWrite.png)
|
||||||
|
|
||||||
|
|
Loading…
Reference in a new issue