Skip to content

Performance Test - Large Object#3

Benchmark LeoFS v1.3.0

Purpose

We check Read/Write Performance of Large File with LeoFS

Issues

Summary

  • Put 500 x 1GB Files with S3Cmd
    • Parallel spawn 16 instances
    • Base File is served through memdisk /dev/shm/
  • During Loading Phase, the throughput is bounded by Disk I/O rate
  • During Reading Phase, decrease in throughput is observed
    • Start with 250MB/s, drop to 150MB/s after 10 minutes
    • Occasional drop to 50MB/s
  • To rule out effect from basho_bench, a simple bash script is used to verify
    • Same situation

  • As page cache is dropped before every test, try to execute once in the middle

    $ sync && echo 3 | sudo tee /proc/sys/vm/drop_caches - At 13:00, drop page cache of all nodes (gateway, storage) - Throughput shows improvement immediately, back to 250MB/s from 150MB/s - Afterwards showing similar trend as the beginning - (As page cache is filled?)

Next Action

  • Research on why throughput is better when page cache is empty (dropped)
  • Detailed breakdown of latency in each processes

Environment

  • OS: Ubuntu Server 14.04.3
  • Erlang/OTP: 17.5
  • LeoFS: 1.3.0
  • CPU: Intel Xeon E5-2630 v3 @ 2.40GHz
  • HDD (node[36~50]) : 4x ST2000LM003 (2TB 5400rpm 32MB) RAID-0 are mounted at /data/, Ext4
  • SSD (node[36~50]) : 1x Crucial CT500BX100SSD1 mounted at /ssd/, Ext4
 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
 [System Confiuration]
-----------------------------------+----------
 Item                              | Value
-----------------------------------+----------
 Basic/Consistency level
-----------------------------------+----------
                    system version | 1.3.0
                        cluster Id | leofs_1
                             DC Id | dc_1
                    Total replicas | 3
          number of successes of R | 1
          number of successes of W | 2
          number of successes of D | 2
 number of rack-awareness replicas | 0
                         ring size | 2^128
-----------------------------------+----------
 Multi DC replication settings
-----------------------------------+----------
        max number of joinable DCs | 2
           number of replicas a DC | 1
-----------------------------------+----------
 Manager RING hash
-----------------------------------+----------
                 current ring-hash | 4adb34e4
                previous ring-hash | 4adb34e4
-----------------------------------+----------

 [State of Node(s)]
-------+------------------------+--------------+----------------+----------------+----------------------------
 type  |          node          |    state     |  current ring  |   prev ring    |          updated at
-------+------------------------+--------------+----------------+----------------+----------------------------
  S    | S1@192.168.100.37      | running      | 4adb34e4       | 4adb34e4       | 2016-10-13 10:29:31 +0900
  S    | S2@192.168.100.38      | running      | 4adb34e4       | 4adb34e4       | 2016-10-13 10:29:31 +0900
  S    | S3@192.168.100.39      | running      | 4adb34e4       | 4adb34e4       | 2016-10-13 10:29:31 +0900
  S    | S4@192.168.100.40      | running      | 4adb34e4       | 4adb34e4       | 2016-10-13 10:29:31 +0900
  S    | S5@192.168.100.41      | running      | 4adb34e4       | 4adb34e4       | 2016-10-13 10:29:31 +0900
  G    | G0@192.168.100.35      | running      | 4adb34e4       | 4adb34e4       | 2016-10-13 10:29:32 +0900
-------+------------------------+--------------+----------------+----------------+----------------------------

Monitoring Results:

  • Loading ops-latency monitoring-results

  • Reading ops-latency monitoring-results

  • Reading (Drop Cache at 1300) monitoring-results