import sys

sys.setrecursionlimit(10**6)
int1 = lambda x: int(x)-1
p2D = lambda x: print(*x, sep="\n")
def II(): return int(sys.stdin.buffer.readline())
def LI(): return list(map(int, sys.stdin.buffer.readline().split()))
def LI1(): return list(map(int1, sys.stdin.buffer.readline().split()))
def LLI(rows_number): return [LI() for _ in range(rows_number)]
def LLI1(rows_number): return [LI1() for _ in range(rows_number)]
def BI(): return sys.stdin.buffer.readline().rstrip()
def SI(): return sys.stdin.buffer.readline().rstrip().decode()
# dij = [(0, 1), (-1, 0), (0, -1), (1, 0)]
dij = [(0, 1), (-1, 0), (0, -1), (1, 0), (1, 1), (1, -1), (-1, 1), (-1, -1)]
inf = 10**16
md = 998244353
# md = 10**9+7

n, k = LI()
if n == k or n%k or n*(n+1)//2%k:
    print("No")
    exit()

ans = [[] for _ in range(k)]
h = n//k

if h & 1:
    for j0 in range(k):
        j = j0
        for i in range(h-3, h):
            ans[j0].append(i*k+j+1)
            j = (j-1)%k
    h -= 3

for a in range(h*k//2):
    ans[a%k] += [a+1, h*k-a]

print("Yes")
for row in ans: print(*row)