Skip to contentSkip to conversation

Comparison of PAE and Non-PAE 2..4.14 (p8) in high load

DISCUSSION Willi Nüßer 2 participants 3 messages Nov 14, 2001

lore
Share & tools
lore
#0 lore raw

Hi,

after my first posting to lkml where we compared distributor provided kernels vs. a plain 2.4.14-pre8 it was pointed out that between PAE and non-PAE kernels some performance differences might exist.

We checked this last night and here are the first results. Again,
a) the relevant quantity dialog steps per second is a measure for the throughput our application servers runs.
b) our application server and the corresponding database (SAP DB) run on 4 way Dell, 1 GB at boot time enabled.

Results:
---------

2.4.7

	2.4.14p8 PAE		2.4.14p4 non- PAE

-------------------------------------------------------------
1.80 13.42 15.47
1.10 13.28 14.76
1.20 14.08 14.63
1.26 13.17 15.30
1.35 13.41 14.51

This means that we did see a performance decrease of about 6 % compared to 2.4.14p8 nonPAE but still 2.4.14p8 is an order of magnitude faster than 2.4.7

Signature · 6 lines
Best regards
	Willi

-----------------------------------
Willi Nuesser
SAP Linuxlab
#1 lore raw
Quoted 27 lines · Willi Nüßer

Hi,

after my first posting to lkml where we compared distributor provided kernels vs. a plain 2.4.14-pre8 it was pointed out that between PAE and non-PAE kernels some performance differences might exist.

We checked this last night and here are the first results. Again,
a) the relevant quantity dialog steps per second is a measure for the throughput our application servers runs.
b) our application server and the corresponding database (SAP DB) run on 4 way Dell, 1 GB at boot time enabled.

Results:
---------

2.4.7

        2.4.14p8 PAE            2.4.14p4 non- PAE
-------------------------------------------------------------
   1.80         13.42                   15.47
   1.10         13.28                   14.76
   1.20                 14.08                   14.63
   1.26         13.17                   15.30
   1.35         13.41                   14.51
Quoted 3 lines

This means that we did see a performance decrease of about 6 % compared to 2.4.14p8 nonPAE but still 2.4.14p8 is an order of magnitude faster than 2.4.7

PAE mode increases the size of the page table entries to 64-bits, and the x86 doesn't do 64-bit operations very well. Plus it has three levels of tables to work with instead of two. It is the only way to support more than 4GB of memory so it's a tradeoff between high memory support and performance. If you have less memory, don't run a PAE kernel.

--

Brian Gerst

#2 lore raw
In reply to Willi Nüßer
Quoted 18 lines · Brian Gerst
Quoted 13 lines

Results:
---------

2.4.7

       2.4.14p8 PAE            2.4.14p4 non- PAE
-------------------------------------------------------------
  1.80         13.42                   15.47
  1.10         13.28                   14.76
  1.20                 14.08                   14.63
  1.26         13.17                   15.30
  1.35         13.41                   14.51
Quoted 4 lines

This means that we did see a performance decrease of about 6 % compared to 2.4.14p8 nonPAE but still 2.4.14p8 is an order of magnitude faster than 2.4.7

Quoted 4 lines

PAE mode increases the size of the page table entries to 64-bits, and the x86 doesn't do 64-bit operations very well. Plus it has three

levels of tables to work with instead of two.

Yes, I know the difference. The reason for this post was not blind
curiosity but the presumption that the great performance increase of 2.4.14p8 we reported was mainly due to the PAE enabling in the 2.4.7 and non-PAE-enabling 2.4.14p8 in our first test.

The tests above show that this is not the case. Even with PAE 2.4.14 is faster by an order of magnitude.

The other info we did find interesting is the actual amount of difference between PAE and nonPAE. The 6% we got fit very well in some estimates I found on lkml.

Signature · 6 lines
Best regards
Willi

-----------------------------------
Willi Nuesser
SAP Linuxlab