First-of-Its-Kind Non-Custodial Exchange Using Confluent Cloud

Nash is using confluent cloud, google cloud platform to deliver and manage its services. Kubernetes and apache kafka technologies will help it scale faster, maintain top notch records, give real time services which are even hard to imagine today.

Look at the customers Confluent apache kafka has

8 Likes

Recently I read jay Hoa(CEO - OkEx) complaining that the Kafka is not a good solution for High Frequency trading and they had to develop their own system to be more responsive to High frequency trading.

Is the team aware to these issues?

1 Like


Checking the original tweet that Jay replied to and the way how he posts more tweets make me think he was just out to get attention from CZ to use for ā€œCZ needs okexā€™s helpā€.

CZ only said there was a problem, and there is no proof that binance uses kafka.
The media went ofcource overboard and made articles how CZ refused help and blocked him- very likely paid articles.

However yeah- would be good if team can say something about this kafka thing, if it will be sufficient for very high trading activity.

1 Like

:popcorn:
.
.
.
Iā€™m trolling but I do find the subject interesting (though Iā€™m pretty convinced the team investigated the question thoroughly before building a big part of their architecture on it).

Lets see what the team has to about thisā€¦

@canesin please comment on this.

1 Like

Apparently they are having some issues with Kafkaā€¦

1 Like

thatā€™s the reason why I wanted a discussion over here.

Wanted to know what the team has thought and how do they aim to tackle issues related to messaging system.

short term and long term.

2 Likes

Use of Kafka per-see is not a issue on our system, it provides the correct guarantees and performance. I donā€™t know how Binance and OKEx are using it.

Short term we are making the application more robust against broker issues, mid term we will guarantee to have full control of operations and long term we will internalize on our own messaging system.

8 Likes

thanks fabio.