Date: Sat, 8 Apr 2023 19:51:26 -0400
On 4/8/23 18:44, Thiago Macieira wrote:
>> Performance and latency are very important factors in the financial
>> sector as the requests need to be processed in real-time.
>
> No dispute here. Performance and latency are important to everyone, actually.
> The only question is the relative priority of those versus other things, like
> complexity / maintainability of the code, memory footprint, time to market,
> availability and cost of developers, of tooling, etc.
>
> But you have to show that there *is* a problem to be solved and show how your
> proposed idea solves it, plus contrast it with other possible solutions. Like
> I said above, improving latency would be welcome by a lot of people; but how
> does your solution do that?
Well I just mentioned the problems (security, efficiency, robustness,
..). Here are the solutions:
1) Well for the security part solved by C++ Superset, here's its
documentation:
https://www.fornux.com/files/pdf/Superset_Manual.pdf
But C++ Superset is patented and commercial so I don't expected to make
it part of the standards but if some corporation would like to partner
up then we can discuss. I already talked to Intel but I put emphasis on
the incomplete Python -> C++ source-to-source compiler. C++ Superset
works perfectly well but parsing all types syntax of existing code,
without requiring modifications, is the part I slightly still need to
fine tune.
2) For low latency FIFO containers we can integrate the following one:
https://github.com/philippeb8/cplusplus-overload-new
It is still incomplete as it lacks a simple subtraction for deallocated
blocks but you get the idea.
It is 3.6 times faster under Linux so I expect an even greater standard
deviation under Windows.
3) I know C++ hash maps are very slow and all companies requiring fast
hash maps replace the standard one. So some attention should be put there.
Even Linus Torvalds would stop complaining C++ is fixing the wrong
problems because he's definitely right.
Thank you,
>> Performance and latency are very important factors in the financial
>> sector as the requests need to be processed in real-time.
>
> No dispute here. Performance and latency are important to everyone, actually.
> The only question is the relative priority of those versus other things, like
> complexity / maintainability of the code, memory footprint, time to market,
> availability and cost of developers, of tooling, etc.
>
> But you have to show that there *is* a problem to be solved and show how your
> proposed idea solves it, plus contrast it with other possible solutions. Like
> I said above, improving latency would be welcome by a lot of people; but how
> does your solution do that?
Well I just mentioned the problems (security, efficiency, robustness,
..). Here are the solutions:
1) Well for the security part solved by C++ Superset, here's its
documentation:
https://www.fornux.com/files/pdf/Superset_Manual.pdf
But C++ Superset is patented and commercial so I don't expected to make
it part of the standards but if some corporation would like to partner
up then we can discuss. I already talked to Intel but I put emphasis on
the incomplete Python -> C++ source-to-source compiler. C++ Superset
works perfectly well but parsing all types syntax of existing code,
without requiring modifications, is the part I slightly still need to
fine tune.
2) For low latency FIFO containers we can integrate the following one:
https://github.com/philippeb8/cplusplus-overload-new
It is still incomplete as it lacks a simple subtraction for deallocated
blocks but you get the idea.
It is 3.6 times faster under Linux so I expect an even greater standard
deviation under Windows.
3) I know C++ hash maps are very slow and all companies requiring fast
hash maps replace the standard one. So some attention should be put there.
Even Linus Torvalds would stop complaining C++ is fixing the wrong
problems because he's definitely right.
Thank you,
-- Logo <https://www.fornux.com/> *Phil Bouchard* facebook icon <https://www.linkedin.com/in/phil-bouchard-5723a910/> CTO T: (819) 328-4743 E: phil_at_[hidden]| www.fornux.com <http://www.fornux.com> 8 rue de la Baie| Gatineau (Qc), J8T 3H3 Canada Banner <https://goglobalawards.org/> Le message ci-dessus, ainsi que les documents l'accompagnant, sont destinés uniquement aux personnes identifiées et peuvent contenir des informations privilégiées, confidentielles ou ne pouvant être divulguées. Si vous avez reçu ce message par erreur, veuillez le détruire. This communication (and/or the attachments) is intended for named recipients only and may contain privileged or confidential information which is not to be disclosed. If you received this communication by mistake please destroy all copies.
Received on 2023-04-08 23:51:28