Closed lalithsuresh closed 1 year ago
Merging #346 (2ad1daa) into main (832ce0c) will decrease coverage by
6.14%
. The diff coverage is60.05%
.
name | main~69 [kOp/s] | PR [kOp/s] | Tput change [%] | Assessment | Peak RSS diff |
---|---|---|---|---|---|
q0 | 5478.79 | 5423.49 | -1 | :heavy_check_mark: | -8.8 MB |
q1 | 5398.75 | 5472.79 | 1 | :heavy_check_mark: | 16.1 MB |
q2 | 5630.96 | 5093.66 | -10 | :small_red_triangle_down: | 5.8 MB |
q3 | 5504.96 | 5516.72 | 0 | :heavy_check_mark: | -5.5 MB |
q4 | 3341.37 | 3412.43 | 2 | :heavy_check_mark: | -17.7 MB |
q5 | 5543.82 | 5425.88 | -2 | :heavy_check_mark: | -17.7 MB |
q6 | 3697.7 | 3576.17 | -3 | :heavy_check_mark: | 12.5 MB |
q7 | 2908 | 4037.55 | 39 | :evergreen_tree: | -4.0 GB |
q8 | 5260.06 | 5216.27 | -1 | :heavy_check_mark: | -4.0 GB |
q9 | 627.889 | 593.382 | -5 | :small_red_triangle_down: | -51.1 MB |
q12 | 4913.17 | 4887.92 | -1 | :heavy_check_mark: | -51.1 MB |
q13 | 2697.28 | 3488.47 | 29 | :evergreen_tree: | -51.1 MB |
q14 | 5659.87 | 5426.57 | -4 | :heavy_check_mark: | -51.1 MB |
q15 | 5302.63 | 5316.23 | 0 | :heavy_check_mark: | -51.1 MB |
q16 | 1038.06 | 1041.78 | 0 | :heavy_check_mark: | -51.1 MB |
q17 | 3140.32 | 3058.65 | -3 | :heavy_check_mark: | -51.1 MB |
q18 | 871.812 | 872.995 | 0 | :heavy_check_mark: | -48.4 MB |
q19 | 1025.09 | 1033.82 | 1 | :heavy_check_mark: | -48.4 MB |
q20 | 1069.96 | 1057.46 | -1 | :heavy_check_mark: | -48.4 MB |
q21 | 5468.41 | 5329.35 | -3 | :heavy_check_mark: | -48.4 MB |
q22 | 5234.71 | 5502.03 | 5 | :evergreen_tree: | -48.4 MB |
name | main~69 [s] | PR [s] | Runtime change [%] | Assessment |
---|---|---|---|---|
galen | 27.7409 | 28.3286 | 2 | :heavy_check_mark: |
algorithm | dataset | threads | main~69 [kEVPS] | PR [kEVPS] | Tput change [%] | Assessment | Peak RSS diff |
---|---|---|---|---|---|---|---|
bfs | graph500-22 | 1 | 1755.93 | 1758.48 | 0 | :heavy_check_mark: | 32.8 kB |
bfs | datagen-8_4-fb | 6 | 7705.23 | 7860.25 | 2 | :heavy_check_mark: | 116.5 MB |
pagerank | graph500-22 | 1 | 688.825 | 687.847 | 0 | :heavy_check_mark: | 122.9 kB |
pagerank | datagen-8_4-fb | 6 | 2050.74 | 2026.43 | -1 | :heavy_check_mark: | 203.9 MB |
name | main~69 [kOp/s] | PR [kOp/s] | Tput change [%] | PR DRAM [kOp/s] | DRAM diff [%] | Assessment |
---|---|---|---|---|---|---|
q0 | 2374.54 | 2400.44 | 1 | 2405.05 | 0 | :heavy_check_mark: |
q1 | 1682.87 | 1680.01 | 0 | 1722.01 | -2 | :heavy_check_mark: |
q2 | 2357.63 | 2427.89 | 3 | 2394.18 | 1 | :heavy_check_mark: |
q3 | 1989.03 | 2009.91 | 1 | 2301.37 | -13 | :heavy_check_mark: |
q4 | 360.36 | 365.522 | 1 | 1386.66 | -74 | :heavy_check_mark: |
q5 | 2006.19 | 2006.74 | 0 | 2284.6 | -12 | :heavy_check_mark: |
q6 | 333.035 | 340.373 | 2 | 1398.54 | -76 | :heavy_check_mark: |
q7 | 258.051 | 254.223 | -1 | 1270.78 | -80 | :heavy_check_mark: |
q8 | 2067 | 2113.83 | 2 | 2227.09 | -5 | :heavy_check_mark: |
q9 | 32.9563 | 33.3701 | 1 | 388.776 | -91 | :heavy_check_mark: |
q12 | 829.501 | 848.793 | 2 | 1798.87 | -53 | :heavy_check_mark: |
q13 | 440.688 | 443.307 | 1 | 983.104 | -55 | :heavy_check_mark: |
q14 | 1701.72 | 1729.34 | 2 | 1707.84 | 1 | :heavy_check_mark: |
q15 | 196.648 | 196.718 | 0 | 1189.38 | -83 | :heavy_check_mark: |
q16 | 26.8666 | 27.1151 | 1 | 288.721 | -91 | :heavy_check_mark: |
q17 | 80.9823 | 80.9857 | 0 | 800.009 | -90 | :heavy_check_mark: |
q18 | 60.4819 | 61.3366 | 1 | 803.282 | -92 | :heavy_check_mark: |
q19 | 67.9337 | 68.4015 | 1 | 660.944 | -90 | :heavy_check_mark: |
q20 | 169.217 | 168.369 | -1 | 941.201 | -82 | :heavy_check_mark: |
q21 | 1479.42 | 1533.24 | 4 | 1506.66 | 2 | :heavy_check_mark: |
q22 | 2034.39 | 2016.39 | -1 | 2018.11 | 0 | :heavy_check_mark: |
This patch includes:
Schema changes to work against both Sqlite and Postgres. In particular, we use
TEXT
instead ofVARCHAR
because the sqlx Any driver cannot decodeVARCHAR
responses from Postgres. For incrementing IDs, we useAUTOINCREMENT
in Sqlite andSERIAL PRIMARY KEY
in Postgres.Add parameterized tests for exercising the same tests against both DBs. The Postgres tests needs a Postgres instance to be deployed and accessible, so we use an environment variable RUN_POSTGRES_TESTS to be set to 1 for the Postgres tests to be run. Otherwise, those tests are ignored.
To connect to Postgres, we can build on Postgres' supported environment variables, like PGHOST, PGUSER and PGPASSWORD. This makes it so that we don't have to hard code any of these in our connection strings.