I’m about to migrate a huge monolithic web application and it turns out that some of the approaches I’m thinking to use are anti-patterns 🙂

I cannot see how to extract, for instance, the “trouble ticketing function” into a detached microservice, without connecting to the “Monolith database”?

A small dedicated DB could be created inside the service, but how to keep it in synch with its “parent”, first, and with the ones created in other microservices for shared data like anagraphical data of users and buildings?

Maybe, I thought, I can make calls between the microservice’s APIs to retrieve data, but they cannot be asynchronous to guarantee the consistency of the created ticket…..and here we are with “Synchronous world” 🙂

Do you have some hint on this?

Can you suggest some readings about these problems and how to solve them?

That’s not happen with python as example, if you want the same with java you have to use the wrappers of Numbers like BigDecimal. ]]>

Fib(76) = 3416454622906707 but this formula will compute 3416454622906716 ]]>

public class Fibonacci {

static double sqrt5 = Math.sqrt(5);

static double phi = (1 +sqrt5)/2;

public static long fibonacci(int n) {

return (long) Math.floor((Math.pow(phi, n)/sqrt5 + 0.5));

}

public static void main(String[] args) {

for(int i=0;i “+fibonacci(i));

}

}

]]>in the tabulated version you can remove the

if (val == 0 || val == 1)

return 1;

🙂

]]>I think there’s a typo:

when you solve a DP problem using memoization you do it by maintaining a map if already solved subproblem:

should be

when you solve a DP problem using memoization you do it by maintaining a map OF already solved subproblem:

BTW, very interesting article.

]]>