Cplex

🚀 Installing IBM CPLEX on HPC using Conda

This step-by-step guide helps HPC users install and test IBM CPLEX using Conda and the Miniconda3 module, with a custom environment path.

📦 Prerequisites

• HPC account access
  
  

• A writable home directory (e.g., /home/case_id/)
  
  

• Access to Miniconda3/22.11.1-1 via module load
  
  

🧩 Step 1: Load Miniconda Module

Load Conda from your HPC environment modules:

module load Miniconda3/22.11.1-1

📁 Step 2: Create Conda Environment with CPLEX

Use a custom path with --prefix to install CPLEX and Python 3.10:

conda create --prefix=/home/case_id/cplex python=3.10 cplex -c ibmdecisionoptimization

✅ This installs:

• Python 3.10
  
  

• IBM CPLEX
  
  

• From the official IBM Decision Optimization channel
  
  

🚀 Step 3: Activate the Environment

Because we used a full path, use source activate:

source activate /home/case_id/cplex

✅ Step 4: Verify CPLEX Installation

Run the Python interpreter:

python

Then inside Python:

import cplex

print(cplex.__version__)

If no error appears and a version is printed, you're good to go ✅

🧪 Step 5: Run a Simple Optimization Example (Optional)

Paste this code in Python to test CPLEX functionality:

from cplex import Cplex

model = Cplex()

model.set_problem_type(Cplex.problem_type.LP)

model.set_objective_sense(model.objective.sense.maximize)

model.variables.add(names=["x1", "x2"], obj=[1.0, 2.0])

model.linear_constraints.add(

    lin_expr=[[["x1", "x2"], [1.0, 1.0]], [["x1"], [1.0]]],

    senses=["L", "L"],

    rhs=[1.0, 0.5]

)

model.solve()

print("Status:", model.solution.get_status_string())

print("Objective value:", model.solution.get_objective_value())

print("x1 =", model.solution.get_values("x1"))

print("x2 =", model.solution.get_values("x2"))

📦 Step 6: (Optional) Install docplex for Python Modeling

To work with CPLEX via a higher-level modeling interface:

conda install -c conda-forge docplex

Then test it in Python:

from docplex.mp.model import Model