Thanks Jaikiran. This is is useful. However, as you point out, the ciphers
are an important factor. It would be good to ensure that the encryption
strength is comparable to make it fair. The first step could be to output
the ciphers used by Java 9 and OpenSSL by default.
Ismael
On 11 Nov 2017 7:3
I ran these same tests with Java 9 runtime today and have updated the
blog to include these numbers[1]. I'm pasting the summary for Java 9 here:
- Both for producer and consumer, there's a *drastic improvement in the
JRE shipped SSLEngine numbers, in almost all metrics, in Java 9 as
compared t
I haven't yet had a chance to try out Java 9, but that's definitely on
my TODO list, maybe sometime this weekend.
Thanks for pointing me to KAFKA-2561. I had missed that.
-Jaikiran
On 30/10/17 4:17 PM, Mickael Maison wrote:
Thanks for sharing, very interesting read.
Did you get a chance to
Thanks for sharing, very interesting read.
Did you get a chance to try JDK 9 ?
We also considered using OpenSSL instead of JSSE especially since
Netty made an easy to re-use package (netty-tcnative).
There was KAFKA-2561
(https://issues.apache.org/jira/browse/KAFKA-2561) where people shared
a fe
We have been using Kafka in some of our projects for the past couple of
years. Our experience with Kafka and SSL had shown some performance
issues when we had seriously tested it (which admittedly was around a
year back). Our basic tests did show that things had improved over time
with newer ve
