[Esug-list] Concurrency Best Practices + Tests

Nowak, Helge HNowak at cincom.com
Thu Sep 5 11:14:50 EDT 2019


Adding to my previous speakers …

At a (non-Smalltalk) conference I once watched a very comprehensive and good presentation by a Professor who also worked for companies. So he had both the theoretical and the practical background and he covered a few different uses cases. The closing summary of his presentation was:

-        Concurrency is a must (in the light of the end of Moore’s Law)

-        Avoid concurrency as long as you can

-        If you really need concurrency then there is no one-size-fits-all solution: each performance/scaling problem is different and needs a different approach

I am sorry I don’t remember the neither the name of the Professor nor the name of the conference ☹ Yet if I do and find the presentation on the web I will distribute the link here.

Cheers
Helge

Helge Nowak
Cincom Smalltalk Technical Account Manager
[http://internal.cincom.com/signature/logo.png]<http://www.cincomsmalltalk.com/>
Cincom Systems GmbH & Co. oHG
Humboldtstraße 3
60318 Frankfurt am Main
GERMANY
office
mobile

website
email
+49 89 89 66 44 94
+49 172 74 00 402

http://www.cincomsmalltalk.com
hnowak at cincom.com
A standpoint is an intellectual horizon of radius zero. -- Albert Einstein

Geschäftsführer/Managing Directors: Thomas M. Nies, Donald E. Vick
oHG mit Sitz/based in Frankfurt am Main (Amtsgericht Frankfurt am Main HRA 50297)
Pers. haftender Gesellschafter/Partner liable to unlimited extent:
Cincom Systems Verwaltungsgesellschaft mbH (Amtsgericht Königstein/Ts. HRB 5069)

--- CONFIDENTIALITY STATEMENT ---
This e-mail transmission contains information that is intended to be privileged and confidential. It is intended only for the addressee named above. If you receive this e-mail in error, please do not read, copy or disseminate it in any manner. If you are not the intended recipient, any disclosure, copying, distribution or use of the contents of this information is prohibited, please reply to the message immediately by informing the sender that the message was misdirected. After replying, please erase it from your computer system. Your assistance in correcting this error is appreciated.


From: Esug-list <esug-list-bounces at lists.esug.org> On Behalf Of Maarten Mostert
Sent: Thursday, 5 September 2019 10:57
To: esug-list at lists.esug.org
Subject: Re: [Esug-list] Concurrency Best Practices + Tests

Hello,

Supposing you are working on a desktop app and you need to keep your main image responsive and want to offload things to a worker image on a separate processor you can hook it to the keyboard processor once finished.

If you are basically making a clone of your image it is easy to make a switch (dev/runtime) to do all the development and debugging in the same image.

This is at least my scenario. The main application works on a read-write sqlite file. The worker makes a read only copy , slowly synchronises everything with Postgres on compose.io<http://compose.io>, when finished puts itself on the keyboard queue and when it his turn writes the differences to the local sqlite work file and signals an UI update.

The keyboard hook makes it possible for the worker to come back on the main thread as a normal user interaction.

Hope this helps,

Maarten,



Le 5 sept. 2019 à 10:27, Karsten Kusche <karsten at heeg.de<mailto:karsten at heeg.de>> a écrit :

Precisely my thinking! try to avoid concurrency as much as possible and if you _really_ have to use it, try a queue approach.


Georg Heeg eK

Wallstraße 22

06366 Köthen



Tel.: 03496/214328

FAX: 03496/214712

Amtsgericht Dortmund HRA 12812



Am 5. September 2019 um 09:34:25, Tim Mackinnon (tim at testit.works<mailto:tim at testit.works>) schrieb:

Not to sound flippant - but do everything you can to limit where you use concurrency (and if possible, not at all). It adds lots of complication that often isn’t necessary.

Where you do need it, queue results or commands and then have workers to process those - workers, queues and commands put you back into easily testable territory.

Then you are left with the concurrent bit - hopefully now quite tiny and specific, which you you can normally blast and test if you have concurrency hole - but in the realms of can it queue things up in an orderly fashion and dequeue them as well. Deep code inspection on this is also helpful (albeit manual).

Tim

Sent from my iPhone

On 4 Sep 2019, at 14:55, John M McIntosh <johnmci at smalltalkconsulting.com<mailto:johnmci at smalltalkconsulting.com>> wrote:
This was useful in doing our single threaded JavaScript to Multi-threaded Swift IOS platform code http://reactivex.io<http://reactivex.io/>




....
John M. McIntosh. Corporate Smalltalk Consulting Ltd https://www.linkedin.com/in/smalltalk


Sent from ProtonMail Mobile


On Wed, Sep 4, 2019 at 15:32, N. Bouraqadi <bouraqadi at gmail.com<mailto:bouraqadi at gmail.com>> wrote:
Hi everyone,

Can I get your input on the following questions :

- What are your best practices and recommendations for developing and testing concurrent software?

- How to discover need for synchronization/critical sections/ when doing TDD?

- How to write code to avoid dead-locks?

Thanks for your help,
Noury
_______________________________________________
Esug-list mailing list
Esug-list at lists.esug.org<mailto:Esug-list at lists.esug.org>
http://lists.esug.org/mailman/listinfo/esug-list_lists.esug.org


_______________________________________________
Esug-list mailing list
Esug-list at lists.esug.org<mailto:Esug-list at lists.esug.org>
http://lists.esug.org/mailman/listinfo/esug-list_lists.esug.org
_______________________________________________
Esug-list mailing list
Esug-list at lists.esug.org<mailto:Esug-list at lists.esug.org>
http://lists.esug.org/mailman/listinfo/esug-list_lists.esug.org
_______________________________________________
Esug-list mailing list
Esug-list at lists.esug.org<mailto:Esug-list at lists.esug.org>
http://lists.esug.org/mailman/listinfo/esug-list_lists.esug.org

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.esug.org/pipermail/esug-list_lists.esug.org/attachments/20190905/488c52e8/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image001.png
Type: image/png
Size: 1652 bytes
Desc: image001.png
URL: <http://lists.esug.org/pipermail/esug-list_lists.esug.org/attachments/20190905/488c52e8/attachment-0001.png>


More information about the Esug-list mailing list