Block size and scalability


#1

If I understood well, NEM solves the problem of scalability with a dynamic block size, and each block is just the size of the transactions in it. Is that correct?


#2

No, every crypto currency uses dynamic block sizes in the sense that the more transactions a block contains the bigger the block is. NEM is not special in that sense.


#3

As far as I know, Bitcoin has a 1 MB block size.


#4

That is the max block size, it can be smallet if the miner puts only few transactions into the block.


#5

Thanks. Does NEM have a max block size?


#6

NEM has a max of 120 txes per block which means that currently the network is (artificially) limited to 2 txes/s.


#7

but of course this will be 4000 TX/ S in the near future with catapult.


#8

In the mijin demonstration experiment, the performance of 4000 tx/sec is issued.
But that is not geographically dispersed, it is the value of using 32 GB RAM 8 core CPU etc.
In order to achieve this performance, it is also necessary to separate the current NIS server into a DB server and an API server and operate them.

The public chain NEM does not need the performance so far.
I guess three digits are the limit as a specification that can guarantee operation.


#9

The public chain will never do 4k txes/s. And it is not needed too.


#10

Why will never do 4k TX/S? Isn’t a feature of catapult mainnet upgrade? Saying that will never do 4k TX/S sounds really bad and negative.


#11

First of all there is no need for 4k txes/s in the public network.
Also, to have 4k txes/s you need big servers, you cannot do it with a 2GB vps.
Those 4k txes/s are aimed at private blockchain use cases. So it is interesting for major companies only.


#12

Is it possible to increase this limit or it is pretty much fixed?


#13

Not really fixed but i think it might be smarter to explore the capabilities of catapult since ultimately it will replace NIS and will have a larger block size.


#14

Yes, we’re all waiting for Catapult to go public!