#include <bits/stdc++.h>

using namespace std;

using ll = long long;
using uint = unsigned;

template<int P>
struct mint {
	int v;
	mint (int _v = 0) {v = _v;}
	operator int() {return v;}
	friend mint operator + (mint x , int y){if((x.v += y) >= P) x.v -= P; return x;}
	friend mint operator - (mint x , int y){if((x.v -= y) < 0) x.v += P; return x;}
	friend mint operator * (mint x , int y){return 1ll * x.v * y % P;}
	mint qpow(int y){mint res(1) , t(v); while(y){if(y & 1) res = res * t; t = t * t; y >>= 1;} return res;}
	friend mint operator ~ (mint x){return x.qpow(P - 2);}
	friend mint operator / (mint x , int y){return x * ~mint(y);}
	friend ostream &operator << (ostream &os , mint &x){return os << x.v , os;}
	friend istream &operator >> (istream &is , mint &x){return is >> x.v , is;}
};
using mi = mint<998244353>;

const int N = 1e7 + 5;

mi fac[N] , ifac[N];
mi A(int x , int y){
	return fac[x] * ifac[x - y];
}

int n , k;
mi f[N] , g[N];

signed main() {
#ifdef LOCAL
#elif defined(ONLINE_JUDGE)
#else 	
	freopen("b.in" , "r" , stdin);
	freopen("b.out" , "w" , stdout);
#endif
	
	fac[0] = 1;
	for(int i = 1; i < N; ++ i) fac[i] = fac[i - 1] * i;
	ifac[N - 1] = ~fac[N - 1];
	for(int i = N - 2; i >= 0; -- i) ifac[i] = ifac[i + 1] * (i + 1);
	
	ios::sync_with_stdio(0); cin.tie(0);
	
	cin >> n >> k;
	for(int i = 1; i <= k; ++ i){
		f[i] = A(n - 1 , i - 1);
	}
	
	for(int i = 1; i <= n; ++ i){
		int j = i <= k ? 1 : i - k;
		f[i] = f[i] + (g[i - 1] - g[j - 1]) * ifac[n - i];
		g[i] = g[i - 1] + f[i] * fac[n - i - 1];
	}
	
	cout << f[n];
	
	return 0;
}