[
https://issues.apache.org/jira/browse/CRYPTO-63?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15353477#comment-15353477
]
Sebb commented on CRYPTO-63:
----------------------------
Unfortunately the Benchmarking results are not good:
{code}
Benchmark Mode Cnt Score Error Units
CryptoBenchmark.CipherCreateJce avgt 20 0.005 ± 0.001 ms/op
CryptoBenchmark.CipherCreateOpenssl avgt 20 0.002 ± 0.001 ms/op
CryptoBenchmark.CipherCreateOpensslJna avgt 20 0.005 ± 0.001 ms/op
CryptoBenchmark.CipherTestJce avgt 20 0.013 ± 0.001 ms/op
CryptoBenchmark.CipherTestOpenssl avgt 20 0.007 ± 0.001 ms/op
CryptoBenchmark.CipherTestOpensslJna avgt 20 0.016 ± 0.003 ms/op
CryptoBenchmark.RandomCreateJava avgt 20 0.002 ± 0.001 ms/op
CryptoBenchmark.RandomCreateOS avgt 20 0.552 ± 0.008 ms/op
CryptoBenchmark.RandomCreateOpenssl avgt 20 0.002 ± 0.001 ms/op
CryptoBenchmark.RandomCreateOpensslJNA avgt 20 0.010 ± 0.004 ms/op
CryptoBenchmark.RandomTestJava avgt 20 0.039 ± 0.001 ms/op
CryptoBenchmark.RandomTestOS avgt 20 0.553 ± 0.003 ms/op
CryptoBenchmark.RandomTestOpenssl avgt 20 0.012 ± 0.001 ms/op
CryptoBenchmark.RandomTestOpensslJNA avgt 20 0.034 ± 0.006 ms/op
{code}
The JNA version takes 2-3 times as long as the native version.
Its Cipher is even slower than the JCE version.
However its Random implementation is slightly faster than the Java version.
Maybe there are ways to speed JNA up.
> Add JNA binding
> ---------------
>
> Key: CRYPTO-63
> URL: https://issues.apache.org/jira/browse/CRYPTO-63
> Project: Commons Crypto
> Issue Type: Improvement
> Reporter: Hendrik Saly
> Assignee: Hendrik Saly
>
> Needs benchmarking before merge.
> PR https://github.com/apache/commons-crypto/pull/47
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
