summaryrefslogtreecommitdiff
path: root/README.org
diff options
context:
space:
mode:
authorJohn Turner <jturner.usa@gmail.com>2025-09-14 20:54:49 -0400
committerJohn Turner <jturner.usa@gmail.com>2025-09-14 20:54:49 -0400
commit60133578815a3611ad0871c3f79b8b0185e0585c (patch)
treec41bb1e50224cebe44e99b0b28fa3eaf68cf81d2 /README.org
parent9d5cc6f8b6ecf75bd8673590a4da073df140f02d (diff)
downloadsqlite-kv-bench-60133578815a3611ad0871c3f79b8b0185e0585c.tar.gz
change the benchmark executable to slurp keys from the database instead of stdin
Diffstat (limited to 'README.org')
-rw-r--r--README.org4
1 files changed, 1 insertions, 3 deletions
diff --git a/README.org b/README.org
index 4f8f5ae..3773bae 100644
--- a/README.org
+++ b/README.org
@@ -1,10 +1,8 @@
* how to use
First create the test database with ~scripts/mkdb.py~.
-The benchmark executable reads keys from stdin.
-
An example run of the benchmark:
#+BEGIN_SRC_BASH
-sqlite3 "${database}" "select kv.key from kv" | shuf | bench ${database}
+bench ${database} ${limit:-10000} ${seed:-${RANDOM}}
#+END_SRC