Readdir ls -U graphs

From Linux NFS

Revision as of 18:21, 7 April 2011 (view source) Amschuma (Talk | contribs) (→NFS v3) ← Older edit		Revision as of 18:24, 7 April 2011 (view source) Amschuma (Talk | contribs) (→NFS v3) Newer edit →
Line 30:		Line 30:
	=== NFS v3 ===		=== NFS v3 ===
	* [[File:v3_ls_U_noplus_rpc.jpg]]		* [[File:v3_ls_U_noplus_rpc.jpg]]
		+	** 2.6.36 uses fewer RPC operations than 2.6.38
	* [[File:v3_ls_U_plus_rpc.jpg]]		* [[File:v3_ls_U_plus_rpc.jpg]]
		+	** Kernels using readdir for large directories issue fewer RPC operations than kernels using readdir plus
		+
	=== NFS v4 ===		=== NFS v4 ===
	* [[File:v4_ls_U_noplus_rpc.jpg]]		* [[File:v4_ls_U_noplus_rpc.jpg]]
	* [[File:v4_ls_U_plus_rpc.jpg]]		* [[File:v4_ls_U_plus_rpc.jpg]]

---

Revision as of 18:24, 7 April 2011

    36: Linux 2.6.36
36NORC: Linux 2.6.36 (without readdir cap)
    38: Linux 2.6.38
  38RC: Linux 2.6.38 (with readdir cap added back)
  38NB: Linux 2.6.38 (with Neil Brown's patch and loop detection)

Contents 1 Real 1.1 NFS v3 1.2 NFS v4 2 Sys 2.1 NFS v3 2.2 NFS v4 3 RPC ops 3.1 NFS v3 3.2 NFS v4

Real

NFS v3

• 
• 
  • In this case, using readdir plus increases the time required for the command to complete.

NFS v4

• 
• 

Sys

NFS v3

• 
• 
  • Using readdir plus causes the command to spend more time in the kernel

NFS v4

• 
• 

RPC ops

NFS v3

• 
  • 2.6.36 uses fewer RPC operations than 2.6.38
• 
  • Kernels using readdir for large directories issue fewer RPC operations than kernels using readdir plus

NFS v4

• 
•