diff --git a/Cargo.lock b/Cargo.lock
index 714ec9e..16f2d71 100644
--- a/Cargo.lock
+++ b/Cargo.lock
@@ -22,6 +22,8 @@ version = "0.1.0"
 dependencies = [
  "benchlib-rs",
  "crossbeam-channel",
+ "crossbeam-utils",
+ "num_cpus",
  "rayon",
 ]
 
diff --git a/Cargo.toml b/Cargo.toml
index 576c0ff..452fe08 100644
--- a/Cargo.toml
+++ b/Cargo.toml
@@ -13,3 +13,5 @@ opt-level = 1
 rayon = "1.3.0"
 crossbeam-channel = "0.4.2"
 benchlib-rs = "0.2.2"
+crossbeam-utils = "0.7.2"
+num_cpus = "1.0"
\ No newline at end of file
diff --git a/src/main.rs b/src/main.rs
index 6544b6e..1efffaa 100644
--- a/src/main.rs
+++ b/src/main.rs
@@ -7,16 +7,25 @@ use benchlib::benching::Bencher;
 pub fn main() {
     let mut bencher = Bencher::new();
     bencher
-        .set_iterations(1000000)
+        .set_iterations(10000)
         .print_settings()
         .bench("Dry Run", || {})
         .bench("Multiply to 100", || to_bench::multiply_to(100))
         .bench("Summation from 0u128 to 1000", || {
-            to_bench::summation_to_1000()
+            to_bench::summation_to(1000)
         })
+        .bench(
+            "Parallel summation with arc mutex from 0u128 to 1000",
+            || to_bench::summation_using_mutex(1000),
+        )
+        .compare()
         .set_iterations(1000)
         .print_settings()
-        .bench("Spawn and Stop thread", || to_bench::start_stop_thread())
+        .bench("Spawn and stop thread", || to_bench::start_stop_thread())
+        .bench("Start and stop threads==cpus", || {
+            to_bench::start_and_wait_for_num_cpu_threads()
+        })
+        .compare()
         .bench("MPSC channel transmit 1000x u128", || {
             to_bench::send_mpsc_channel()
         })
@@ -32,3 +41,18 @@ pub fn main() {
         })
         .compare();
 }
+
+#[cfg(test)]
+pub mod test {
+    use crate::to_bench::{largest_prime, largest_prime_par, summation_to, summation_using_mutex};
+
+    #[test]
+    pub fn summation_works() {
+        assert_eq!(summation_to(1000), summation_using_mutex(1000))
+    }
+
+    #[test]
+    pub fn primes_work() {
+        assert_eq!(largest_prime(10000), largest_prime_par(10000))
+    }
+}
diff --git a/src/to_bench.rs b/src/to_bench.rs
index 04dd34b..58b9c1f 100644
--- a/src/to_bench.rs
+++ b/src/to_bench.rs
@@ -1,6 +1,9 @@
 use crossbeam_channel::unbounded;
+use crossbeam_utils::sync::WaitGroup;
+use num_cpus;
 use rayon::prelude::*;
 use std::sync::mpsc::channel;
+use std::sync::{Arc, Mutex};
 use std::thread;
 
 pub fn start_stop_thread() {
@@ -10,9 +13,9 @@ pub fn start_stop_thread() {
     handle.join().unwrap();
 }
 
-pub fn summation_to_1000() -> u128 {
+pub fn summation_to(end: u128) -> u128 {
     let mut res = 0u128;
-    for i in 0u128..1000 {
+    for i in 0u128..end {
         res += i;
     }
     res
@@ -82,3 +85,36 @@ pub fn send_mpmc_channel() {
     std::mem::drop(rx);
     handle.join().unwrap();
 }
+
+pub fn summation_using_mutex(end: usize) -> u128 {
+    let wg = WaitGroup::new();
+    let num = Arc::new(Mutex::new(0u128));
+
+    for thread_number in 0..num_cpus::get() {
+        let wg = wg.clone();
+        let num = Arc::clone(&num);
+        thread::spawn(move || {
+            for i in (thread_number..end).step_by(num_cpus::get()) {
+                *num.lock().unwrap() += i as u128;
+            }
+            std::mem::drop(wg);
+        });
+    }
+    wg.wait();
+    let result = *num.lock().unwrap();
+
+    result
+}
+
+pub fn start_and_wait_for_num_cpu_threads() {
+    let wg = WaitGroup::new();
+
+    for thread_number in 0..num_cpus::get() {
+        let wg = wg.clone();
+        thread::spawn(move || {
+            std::mem::drop(wg);
+            thread_number
+        });
+    }
+    wg.wait();
+}