[??/many] acrypto benchmarks vs cryptoloop vs dm_crypt

Evgeniy Polyakov johnpol at 2ka.mipt.ru
Mon Mar 7 21:37:33 CET 2005


Benchmark: Bonnie++ 1.03.
Machine: 2-way Xeon (1+1HT), 1Gb ram.
Ext2 filesystem over file(mapped using loop(cryptoloop, dm_crypt) 
or bd_fd filter for bd).

Version  @version@      ------Sequential Output------ --Sequential Input- --Random-
                    -Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- --Seeks--
Machine        Size K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP  /sec %CP
acrypto       2500M  5479  74  8122   5  4886   4  5690  73 10713   4  22.0   0
cryptoloop    2500M  5812  71 10437   7  4402   5  7165  92 10763   6  88.3   0
dm_crypt      2500M  6040  90  6747  36  4768   8  5775  66 10161   5  90.6   0
                    ------Sequential Create------ --------Random Create--------
                    -Create-- --Read--- -Delete-- -Create-- --Read--- -Delete--
files:max:min        /sec %CP  /sec %CP  /sec %CP  /sec %CP  /sec %CP  /sec %CP
acrypto          16  1345  99 +++++ +++ +++++ +++  1403  99 +++++ +++  4538 100
cryptoloop       16  1372  99 +++++ +++ +++++ +++  1405  99 +++++ +++  4501  99
dm_crypt         16  1352  99 +++++ +++ +++++ +++  1371  99 +++++ +++  4278 100


bd+acrypto works exactly as cryptoloop (attitude of the performance 
acrypto vs. cryptoloop is always the same as CPU usage attitude, 
BUT!, I can not setup bd+acrypto to use the same amount of CPU as loopdev!,
so in absolute numbers, cryptoloop is faster).
dm_crypt is slower.




More information about the CryptoAPI mailing list