Java / C # Why does one function with the same parameters show different execution times?

While writing the program I had a question, the answer to which I could not find. I measure the execution time of a single function in nanoseconds with the same parameters and I get not the same result (in Java):

-- 0,000087036 -- 0,000084647 -- 0,000082940 -- 0,000101713 -- 0,000085330 -- 0,000087377 -- 0,000082598 -- 0,000085330 -- 0,000081916 -- 0,000084306 -- 0,000081916 -- 0,000082940 -- 0,000083622 -- 0,000083964 -- 0,000081916 -- 0,000081575 -- 0,000095227 -- 0,000081916 -- 0,000087036 -- 0,000081575 Среднее значение -- 0,000085244

I tried to do the same in C #, there are also different values. Perhaps it is obvious that they cannot coincide due to the very high accuracy of time, but I would like to know the specific reason. How can this be explained? What does it affect? Java code:

 public class MainFloat { final static int RAND_MAX = 32767; public static void main(String[] args) { int[] check = new int[] {1000, 5000, 10000, 20000, 30000, 40000, 50000, 60000, 70000, 80000, 90000, 100000, 200000, 300000, 400000, 500000, 600000, 700000, 800000, 900000, 1000000, 2500000, 5000000, 10000000, 15000000, 20000000}; double counter = 0; double timeSpent = 0; for(int i = 0; i < 26; i++) { int N = check[i]; float[] a = new float[N]; float[] b = new float[N]; float[] c = new float[N]; float[] f = new float[N]; float[] x = new float[N]; for(int h = 1; h <= N - 1; h++) { do { a[h] = (float) ((h == 1) ? 0 : Math.random() * RAND_MAX); b[h] = (float) ((h == N - 1) ? 0 : Math.random() * RAND_MAX); c[h] = (float) (Math.random() * RAND_MAX); } while(Math.abs(c[h]) < Math.abs(a[h]) + Math.abs(b[h])); f[h] = (float) (Math.random() * RAND_MAX); } counter = 0; for(int j = 0; j < 20; j++) { double startTime = System.nanoTime(); Progonka(check[i] - 1, a, b, c, f, x); timeSpent = System.nanoTime() - startTime; System.out.printf(" -- %8.9f%n", timeSpent/1000000000); counter += timeSpent/1000000000; } counter /= 20; System.out.printf("%8.9f%n", counter); } System.out.printf("END"); } static void Progonka(int np, float ap[], float bp[], float cp[], float fp[], float xp[]) { float[] alfa = new float[np + 1]; float[] beta = new float[np + 1]; int ip; for(ip = 1; ip < np; ip++) { alfa[ip + 1] = bp[ip] / (cp[ip] - ap[ip] * alfa[ip]); beta[ip + 1] = (fp[ip] + ap[ip] * beta[ip]) / (cp[ip] - ap[ip] * alfa[ip]); } xp[np] = (fp[np] + ap[np] * beta[np]) / (cp[np] - ap[np] - alfa[np]); for(ip = np - 1; ip >= 1; ip--) xp[ip] = alfa[ip + 1] * xp[ip + 1] + beta[ip + 1]; } 

}