git-annex/Database/Benchmark.hs
Joey Hess 793ddecd4b
use openTempFile from file-io
And follow-on changes.

Note that relatedTemplate was changed to operate on a RawFilePath, and
so when it counts the length, it is now the number of bytes, not the
number of code points. This will just make it truncate shorter strings
in some cases, the truncation is still unicode aware.

When not building with the OsPath flag, toOsPath . fromRawFilePath and
fromRawFilePath . fromOsPath do extra conversions back and forth between
String and ByteString. That overhead could be avoided, but that's the
non-optimised build mode, so didn't bother.

Sponsored-by: unqueued
2025-01-22 11:41:43 -04:00

120 lines
3.7 KiB
Haskell

{- git-annex database benchmarks
-
- Copyright 2016-2019 Joey Hess <id@joeyh.name>
-
- Licensed under the GNU AGPL version 3 or higher.
-}
{-# LANGUAGE CPP #-}
{-# LANGUAGE OverloadedStrings #-}
module Database.Benchmark (benchmarkDbs) where
import Annex.Common
import Types.Benchmark
#ifdef WITH_BENCHMARK
import qualified Database.Keys.SQL as SQL
import qualified Database.Queue as H
import Database.Init
import Utility.Tmp.Dir
import Git.FilePath
import Types.Key
import Utility.DataUnits
import Criterion.Main
import qualified Data.ByteString.Short as S (toShort)
import qualified Data.ByteString.Char8 as B8
import System.Random
import Control.Concurrent
import qualified System.FilePath.ByteString as P
#endif
benchmarkDbs :: CriterionMode -> Integer -> Annex ()
#ifdef WITH_BENCHMARK
benchmarkDbs mode n = withTmpDirIn "." (toOsPath "benchmark") $ \tmpdir -> do
db <- benchDb (toRawFilePath tmpdir) n
liftIO $ runMode mode
[ bgroup "keys database"
[ getAssociatedFilesHitBench db
, getAssociatedFilesMissBench db
, getAssociatedKeyHitBench db
, getAssociatedKeyMissBench db
, addAssociatedFileOldBench db
, addAssociatedFileNewBench db
]
]
#else
benchmarkDbs _ = giveup "not built with criterion, cannot benchmark"
#endif
#ifdef WITH_BENCHMARK
getAssociatedFilesHitBench :: BenchDb -> Benchmark
getAssociatedFilesHitBench (BenchDb h num _) = bench ("getAssociatedFiles (hit)") $ nfIO $ do
n <- getStdRandom (randomR (1,num))
SQL.getAssociatedFiles (keyN n) (SQL.ReadHandle h)
getAssociatedFilesMissBench :: BenchDb -> Benchmark
getAssociatedFilesMissBench (BenchDb h _ _) = bench ("getAssociatedFiles (miss)") $ nfIO $
SQL.getAssociatedFiles keyMiss (SQL.ReadHandle h)
getAssociatedKeyHitBench :: BenchDb -> Benchmark
getAssociatedKeyHitBench (BenchDb h num _) = bench ("getAssociatedKey (hit)") $ nfIO $ do
n <- getStdRandom (randomR (1,num))
SQL.getAssociatedKey (fileN n) (SQL.ReadHandle h)
getAssociatedKeyMissBench :: BenchDb -> Benchmark
getAssociatedKeyMissBench (BenchDb h _ _) = bench ("getAssociatedKey (miss)") $ nfIO $
SQL.getAssociatedKey fileMiss (SQL.ReadHandle h)
addAssociatedFileOldBench :: BenchDb -> Benchmark
addAssociatedFileOldBench (BenchDb h num _) = bench ("addAssociatedFile to (old)") $ nfIO $ do
n <- getStdRandom (randomR (1,num))
SQL.addAssociatedFile (keyN n) (fileN n) (SQL.WriteHandle h)
H.flushDbQueue h
addAssociatedFileNewBench :: BenchDb -> Benchmark
addAssociatedFileNewBench (BenchDb h num mv) = bench ("addAssociatedFile to (new)") $ nfIO $ do
n <- takeMVar mv
putMVar mv (n+1)
SQL.addAssociatedFile (keyN n) (fileN (num+n)) (SQL.WriteHandle h)
H.flushDbQueue h
populateAssociatedFiles :: H.DbQueue -> Integer -> IO ()
populateAssociatedFiles h num = do
forM_ [1..num] $ \n ->
SQL.addAssociatedFile (keyN n) (fileN n) (SQL.WriteHandle h)
H.flushDbQueue h
keyN :: Integer -> Key
keyN n = mkKey $ \k -> k
{ keyName = S.toShort (B8.pack $ "key" ++ show n)
, keyVariety = OtherKey "BENCH"
}
fileN :: Integer -> TopFilePath
fileN n = asTopFilePath (toRawFilePath ("file" ++ show n))
keyMiss :: Key
keyMiss = keyN 0 -- 0 is never stored
fileMiss :: TopFilePath
fileMiss = fileN 0 -- 0 is never stored
data BenchDb = BenchDb H.DbQueue Integer (MVar Integer)
benchDb :: RawFilePath -> Integer -> Annex BenchDb
benchDb tmpdir num = do
liftIO $ putStrLn $ "setting up database with " ++ show num ++ " items"
initDb db SQL.createTables
h <- liftIO $ H.openDbQueue db SQL.containedTable
liftIO $ populateAssociatedFiles h num
sz <- liftIO $ getFileSize db
liftIO $ putStrLn $ "size of database on disk: " ++
roughSize storageUnits False sz
mv <- liftIO $ newMVar 1
return (BenchDb h num mv)
where
db = tmpdir P.</> toRawFilePath (show num </> "db")
#endif /* WITH_BENCHMARK */