gitlab-com-seaweedfs/seaweedfs
1
0
Fork 0
mirror of https://github.com/seaweedfs/seaweedfs.git synced 2024-01-19 02:48:24 +00:00
Code Issues Projects Releases Wiki Activity

Updated FUSE Mount (markdown)

Chris Lu 2020-10-24 23:06:01 -07:00
parent f7566553fb
commit 788cb65cbf
1 changed files with 43 additions and 32 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

75
FUSE-Mount.md

@ -99,16 +99,18 @@ So if your data is temporary local files, try to move the writes to other unmoun
For example, you can create a soft link to a directory or a file on a local disk, and put temp data there.
#### Benchmark Results
#### Sysbench Benchmark Results
"sysbench" is used here. If you have better benchmarking tools, please share your results.
"sysbench" is used here. The mount command line is `weed mount -dir=xx -chunkSizeLimitMB=4`
If you have better benchmarking tools, please share your results.
```
$ brew install sysbench
$ cd /a/mounted/folder
$ sysbench --test=fileio --file-total-size=1G prepare
$ sysbench --test=fileio --file-total-size=1G --file-test-mode=rndrw --max-time=300 --max-requests=0 --num-threads=1 run
$ sysbench --test=fileio --file-total-size=1G --file-test-mode=rndrw --max-time=60 --max-requests=0 --num-threads=1 run
WARNING: the --test option is deprecated. You can pass a script name or path on the command line without any options.
WARNING: --max-time is deprecated, use --time instead
sysbench 1.0.19 (using bundled LuaJIT 2.1.0-beta2)
@ -134,40 +136,40 @@ Threads started!
File operations:
reads/s: 6854.90
writes/s: 4569.93
fsyncs/s: 14624.11
reads/s: 8502.51
writes/s: 5668.35
fsyncs/s: 18140.64
Throughput:
read, MiB/s: 107.11
written, MiB/s: 71.41
read, MiB/s: 132.85
written, MiB/s: 88.57
General statistics:
total time: 300.0030s
total number of events: 7814670
total time: 60.0026s
total number of events: 1938693
Latency (ms):
min: 0.00
avg: 0.04
max: 244.85
95th percentile: 0.15
sum: 292121.06
avg: 0.03
max: 4.32
95th percentile: 0.11
sum: 58311.71
Threads fairness:
events (avg/stddev): 7814670.0000/0.00
execution time (avg/stddev): 292.1211/0.00
events (avg/stddev): 1938693.0000/0.00
execution time (avg/stddev): 58.3117/0.00
```
The above is single-threaded. The following uses 32 threads.
The above is single-threaded. The following uses 16 threads.
```
$ sysbench --test=fileio --file-total-size=1G --file-test-mode=rndrw --max-time=300 --max-requests=0 --num-threads=32 run
$ sysbench --test=fileio --file-total-size=1G --file-test-mode=rndrw --max-time=60 --max-requests=0 --num-threads=16 run
WARNING: the --test option is deprecated. You can pass a script name or path on the command line without any options.
WARNING: --num-threads is deprecated, use --threads instead
WARNING: --max-time is deprecated, use --time instead
sysbench 1.0.19 (using bundled LuaJIT 2.1.0-beta2)
Running the test with following options:
Number of threads: 32
Number of threads: 16
Initializing random number generator from current time
@ -187,31 +189,40 @@ Threads started!
File operations:
reads/s: 18479.45
writes/s: 12319.70
fsyncs/s: 39436.17
reads/s: 20368.01
writes/s: 13578.67
fsyncs/s: 43485.34
Throughput:
read, MiB/s: 288.74
written, MiB/s: 192.50
read, MiB/s: 318.25
written, MiB/s: 212.17
General statistics:
total time: 300.0393s
total number of events: 21069361
total time: 60.0111s
total number of events: 4644841
Latency (ms):
min: 0.00
avg: 0.45
max: 123.55
95th percentile: 1.64
sum: 9547894.18
avg: 0.20
max: 8.22
95th percentile: 0.90
sum: 944779.04
Threads fairness:
events (avg/stddev): 658417.5312/1098.35
execution time (avg/stddev): 298.3717/0.02
events (avg/stddev): 290302.5625/1954.88
execution time (avg/stddev): 59.0487/0.01
```
##### Sysbench Result Analysis
sysbench works on 128 files, 8MiB each. It will do random read and write on these files.
`weed mount` has default `cacheCapacityMB=1000`, but because the cache has different section for different sized chunks, the actual cache used for this workload is about 500MB. However, due to the randomness, the actual hit rate is not high.
Even with the cache, the data are persisted on the filer and volume servers first, then cached locally.
#### FIO Benchmark
Here is the result of using tool [fio](https://github.com/axboe/fio).
* The server and mount are restarted before each run.
* Mount disabled the local cache: `weed mount -dir=xx -cacheCapacityMB=0`