Kernel commit c128e575514c ("NFS: Optimise the default readahead size")
changed the calculation for NFS readahead from a multiple of rsize to
the system default, 128 kiB. This setting has been causing read heavy
workloads to underperform by at least 30%, as show below.
$ cat /sys/class/bdi/0\:55/read_ahead_kb
128
$ for i in {0..3} ; do dd if=/mnt/testfile.bin of=/dev/null bs=1M 2>&1 \
| grep copied ; echo 3 > /proc/sys/vm/drop_caches ; done
4294967296 bytes (4.3 GB, 4.0 GiB) copied, 17.056 s, 252 MB/s
4294967296 bytes (4.3 GB, 4.0 GiB) copied, 17.1258 s, 251 MB/s
4294967296 bytes (4.3 GB, 4.0 GiB) copied, 16.5981 s, 259 MB/s
4294967296 bytes (4.3 GB, 4.0 GiB) copied, 16.5487 s, 260 MB/s
$ echo 15360 > /sys/class/bdi/0\:55/read_ahead_kb
$ for i in {0..3} ; do dd if=/mnt/testfile.bin of=/dev/null bs=1M 2>&1 \
| grep copied ; echo 3 > /proc/sys/vm/drop_caches ; done
4294967296 bytes (4.3 GB, 4.0 GiB) copied, 12.3855 s, 347 MB/s
4294967296 bytes (4.3 GB, 4.0 GiB) copied, 11.2528 s, 382 MB/s
4294967296 bytes (4.3 GB, 4.0 GiB) copied, 11.9849 s, 358 MB/s
4294967296 bytes (4.3 GB, 4.0 GiB) copied, 11.2953 s, 380 MB/s
This patch and the following create a tool for automatically setting NFS
readahead during the mount, nfsrahead. The tool is invoked from udev
when the NFS backing device is created in kernel, and sets the readahead
using the sysfs interface.
Bugzilla: https://bugzilla.redhat.com/show_bug.cgi?id=1946283
Signed-off-by: Thiago Becker <[email protected]>
---
.gitignore | 1 +
configure.ac | 1 +
tools/Makefile.am | 2 +-
tools/nfsrahead/Makefile.am | 3 +++
tools/nfsrahead/main.c | 7 +++++++
5 files changed, 13 insertions(+), 1 deletion(-)
create mode 100644 tools/nfsrahead/Makefile.am
create mode 100644 tools/nfsrahead/main.c
diff --git a/.gitignore b/.gitignore
index c89d1cd2..38ab1d39 100644
--- a/.gitignore
+++ b/.gitignore
@@ -61,6 +61,7 @@ utils/statd/statd
tools/locktest/testlk
tools/getiversion/getiversion
tools/nfsconf/nfsconf
+tools/nfsrahead/nfsrahead
support/export/mount.h
support/export/mount_clnt.c
support/export/mount_xdr.c
diff --git a/configure.ac b/configure.ac
index e0f5a930..3e1c183b 100644
--- a/configure.ac
+++ b/configure.ac
@@ -737,6 +737,7 @@ AC_CONFIG_FILES([
tools/rpcgen/Makefile
tools/mountstats/Makefile
tools/nfs-iostat/Makefile
+ tools/nfsrahead/Makefile
tools/rpcctl/Makefile
tools/nfsdclnts/Makefile
tools/nfsconf/Makefile
diff --git a/tools/Makefile.am b/tools/Makefile.am
index c3feabbe..40c17c37 100644
--- a/tools/Makefile.am
+++ b/tools/Makefile.am
@@ -12,6 +12,6 @@ if CONFIG_NFSDCLD
OPTDIRS += nfsdclddb
endif
-SUBDIRS = locktest rpcdebug nlmtest mountstats nfs-iostat rpcctl nfsdclnts $(OPTDIRS)
+SUBDIRS = locktest rpcdebug nlmtest mountstats nfs-iostat rpcctl nfsdclnts nfsrahead $(OPTDIRS)
MAINTAINERCLEANFILES = Makefile.in
diff --git a/tools/nfsrahead/Makefile.am b/tools/nfsrahead/Makefile.am
new file mode 100644
index 00000000..edff7921
--- /dev/null
+++ b/tools/nfsrahead/Makefile.am
@@ -0,0 +1,3 @@
+libexec_PROGRAMS = nfsrahead
+nfsrahead_SOURCES = main.c
+
diff --git a/tools/nfsrahead/main.c b/tools/nfsrahead/main.c
new file mode 100644
index 00000000..0359aced
--- /dev/null
+++ b/tools/nfsrahead/main.c
@@ -0,0 +1,7 @@
+#include <stdio.h>
+
+int main(int argc, char **argv)
+{
+ unsigned int readahead = 128;
+ printf("%d\n", readahead);
+}
--
2.35.1