Hi Florian,

Thanks for sharing! I know there are plenty of points of optimisation we could do to improve these numbers.

One quick thought: have you tried using the Shenandoah garbage collector? That is great for reducing GC pauses, which is often responsible for problematic latency. This is especially true as you increase the heap size (which will be necessary if you're trying to cope with large numbers of parallel requests)

We also very much endorse Gatling [1] as a performance framework, as it allows you to more accurately simulate your real-world application's behaviour, whilst not suffering from some of the statistical/methodological issues of traditional benchmarking software.

Thanks again for sharing, Florian -- great stuff.

Regards,
Marc 

[1] https://gatling.io/ --  it's a free and open source tool.




On Wed, 28 Nov 2018 at 08:37, Volk, Florian <Florian.Volk@scheer-group.com> wrote:

Hello, apiman comunity

I will try to anwser your questions:

  1. Can we deploy APIMAN on pivotal cloud foundry?

We don’t use this.

  1. How to scale the APIMAN?

We put the vertx gateway in a docker container so we can easily scale thne number of gateways if we need more than one.

  1. Hows is the performance of it?

Our product is based on apiman. But we dockerized the whole apiman stack.
Recently we made some load tests with some policies. Your can see the results in our forum blog post.

https://forum.e2ebridge.com/forums/topic/api-management-loadtests-results/

I hope our information will help you.

 

Best regards

Florian

 

Von: Marc Savy <marc.savy@redhat.com>
Gesendet: Wednesday, 28 November 2018 01:38
An: Pritam.Lahoti@cognizant.com
Cc: Volk, Florian <Florian.Volk@scheer-group.com>; apiman-user@lists.jboss.org
Betreff: Re: [Apiman-user] APIMAN database query

 

On Wed, 28 Nov 2018 at 00:34, <Pritam.Lahoti@cognizant.com> wrote:

Thanks for the replies Team.

 

I have some of the queries as-

  • Can we deploy APIMAN on pivotal cloud foundry?

 

I've never used Pivotal Cloud Foundry, I'm afraid. Perhaps another community member can answer.

 

  • How to scale the APIMAN?

 

You can add as many gateways as you need with the same configuration, which allows you to scale horizontally

 

 

  • Hows is the performance of it?

You'll have to test for your own use-cases and it's highly dependent upon configuration and which components you use. But, it should be more than enough for most use-cases.

 

The real question here is: what are your requirements? 

  • Any licencing or cost information ?

 

Apiman is a free and open source project.

 

 

Thanks and Regdards,

Pritam Lahoti

From: Volk, Florian <Florian.Volk@scheer-group.com>
Sent: Thursday, November 22, 2018 8:24:31 AM
To: Lahoti, Pritam (Cognizant); apiman-user@lists.jboss.org
Subject: AW: APIMAN database query

 

Hi Pritam,

 

maybe our anwers help you with your decision:

 

1.       We use Elasticsearch. Until now it works very well and reliable. Also ES is very fast if it comes to metrics and analysis.

2.       We made around 3 million requests against the gateway, after that we had have around 600 MB of data in our Elasticseach (with metrics enabled of course.)

 

Regards

Florian

 

Von: apiman-user-bounces@lists.jboss.org <apiman-user-bounces@lists.jboss.org> Im Auftrag von Pritam.Lahoti@cognizant.com
Gesendet: Wednesday, 14 November 2018 12:47
An: apiman-user@lists.jboss.org
Betreff: [Apiman-user] APIMAN database query

 

Hi Team,

 

We are currently evaluating APIMAN as API lifecycle management solution and have some specific queries around -

 

  1. Best Suitable database?
  2. Re-commanded Sizing of database as per application size?

 

Thanks and Regards,

Pritam Lahoti

_______________________________________________
Apiman-user mailing list
Apiman-user@lists.jboss.org
https://lists.jboss.org/mailman/listinfo/apiman-user