Merge pull request #281 from coding-for-reproducible-research/280-update-parallel-computing

Update parallel computing material

Author: liamjberrisford

_toc.yml

@@ -251,10 +251,13 @@ parts:
                   - file: individual_modules/introduction_to_hpc/exploiting_slurm
               - file: individual_modules/section_landing_pages/parallel_computing
                 sections:
-                  - file: individual_modules/parallel_computing/hello_world
-                  - file: individual_modules/parallel_computing/simple_communication
-                  - file: individual_modules/parallel_computing/collective_comms
-                  - file: individual_modules/parallel_computing/parallel_fractal
+                  - file: individual_modules/parallel_computing/architecture_and_concurrency
+                  - file: individual_modules/parallel_computing/multithreading_io
+                  - file: individual_modules/parallel_computing/multiprocessing_fractal
+                  - file: individual_modules/parallel_computing/mpi_hello_world
+                  - file: individual_modules/parallel_computing/mpi_simple_communication
+                  - file: individual_modules/parallel_computing/mpi_collective_comms
+                  - file: individual_modules/parallel_computing/mpi_parallel_fractal
           - file: course_homepages/software_development
             sections:
               - file: individual_modules/section_landing_pages/software_development_best_practice

course_homepages/high_performance_computing.ipynb

@@ -49,7 +49,7 @@
     "\n",
     "[Clickable Link to Self Study Notes](../individual_modules/section_landing_pages/parallel_computing.md)\n",
     "\n",
-    "The course offers an in-depth understanding of distributed and shared-memory parallelism, enabling participants to explain these concepts clearly. Students will learn to write software that runs across multiple processes using MPI and develop code that utilizes multithreading for parallel execution. Additionally, the course focuses on effectively dividing and parallelizing problems, providing hands-on experience in writing and optimizing parallel code."
+    "The course offers an introductory understanding of shared memory and distributed-memory parallelism, enabling participants to explain these concepts clearly. Students will learn to write software that runs across multiple threads and processes using Python's multithreading annd multiprocessing modules. They will also learn how to use message passing on distributed memory systems with MPI. Additionally, the course focuses on effectively dividing and parallelizing problems, providing hands-on experience in writing and optimizing parallel code."
    ]
   }
  ],

individual_modules/parallel_computing/architecture_and_concurrency.ipynb

@@ -0,0 +1,27 @@
+import time
+import requests
+
+
+def download_all_sites(sites):
+    with requests.Session() as session:
+        for url in sites:
+            download_single_site(url, session)
+
+
+def download_single_site(url, session):
+    with session.get(url) as response:
+        print(f"Read {len(response.content)} bytes from {url}")
+
+
+# Create list of site to download
+sites = [
+        "https://www.exeter.ac.uk/research/research-software-and-analytics/",
+        "https://coding-for-reproducible-research.github.io/CfRR_Courses/home_page.html",
+    ] * 100
+
+
+# Report time for downloading all sites
+start_time = time.perf_counter()
+download_all_sites(sites)
+run_time = time.perf_counter() - start_time
+print(f"Downloaded {len(sites)} sites in {run_time} seconds")

individual_modules/parallel_computing/complete_files/io_bound_complete.py

@@ -0,0 +1,36 @@
+import threading
+import time
+from concurrent.futures import ThreadPoolExecutor
+import requests
+
+thread_local = threading.local()
+
+
+def download_all_sites(sites):
+    with ThreadPoolExecutor(max_workers=5) as executor:
+        executor.map(download_single_site, sites)
+
+
+def download_single_site(url):
+    session = get_session_for_thread()
+    with session.get(url) as response:
+        print(f"Read {len(response.content)} bytes from {url}")
+
+
+def get_session_for_thread():
+    if not hasattr(thread_local, "session"):
+        thread_local.session = requests.Session()
+    return thread_local.session
+
+
+# Create list of site to download
+sites = [
+        "https://www.exeter.ac.uk/research/research-software-and-analytics/",
+        "https://coding-for-reproducible-research.github.io/CfRR_Courses/home_page.html",
+    ] * 100
+
+# Report time for downloading all sites
+start_time = time.perf_counter()
+download_all_sites(sites)
+run_time = time.perf_counter() - start_time
+print(f"Downloaded {len(sites)} sites in {run_time} seconds")

individual_modules/parallel_computing/complete_files/collective_comms_complete.py renamed to individual_modules/parallel_computing/complete_files/mpi_collective_comms_complete.py

@@ -0,0 +1,35 @@
+import psutil
+
+def get_cpu_info():
+    """Retrieves CPU specifications and usage."""
+
+    try:
+        # Get CPU count
+        cpu_count = psutil.cpu_count(logical=False) # Get physical cores
+        cores = psutil.cpu_count(logical=True) # Get logical cores
+        cpu_usage = psutil.cpu_percent() # Get overall CPU usage
+        
+        return f"CPU: {cpu_count} physical cores, {cores} logical cores\nCPU Usage: {cpu_usage}%"
+
+    except Exception as e:
+        print(f"Error getting CPU info: {e}")
+        return None
+
+def get_ram_info():
+    """Retrieves memory information."""
+
+    try:
+        memory_info = psutil.virtual_memory() # Get virtual memory usage
+        total_memory_gb = memory_info.total / (1024**3) # Convert to GB
+        available_memory_gb = memory_info.available / (1024**3) # Convert to GB
+        used_memory_gb = memory_info.used / (1024**3) # Convert to GB
+
+        return f"RAM: Total {total_memory_gb:.2f} GB, Available {available_memory_gb:.2f} GB, Used {used_memory_gb:.2f} GB"
+
+    except Exception as e:
+        print(f"Error getting RAM info: {e}")
+        return None
+
+print(get_cpu_info())
+print(get_ram_info())
+