Update 01-parallel-programming.md

Author: csccva

intro-day/docs/01-parallel-programming.md

@@ -239,6 +239,25 @@ Simple domains with different computational cost
 </div>
 
 
+# Current Supercomputers
+
+- Large computer clusters
+- A lot of standard, high-end server hardware connected to each other
+- High-speed interconnect between compute nodes
+<div class=column>
+- Example: LUMI
+  - 2064 CPU nodes<br>(~260,000 CPU cores in total)
+  - 2978 GPU nodes<br>(~12,000 MI250X GPUs in total) 
+</div>
+<div class=column>
+<center>
+![](img/lumi2.jpg){.center width=80%}
+<small> EuroHPC JU LUMI (HPE Cray EX) at CSC (2021–) </small>
+</center>
+</div>
+
+
+
 # Lumi
 
  ![](img/lumi.png){.center width=50%}