Faster than Fast

For large or complex models, distributed optimization can dramatically reduce solve times —up to 15× faster with 32 machines, and 2-3x speeds-ups are common with eight. Gurobi offers three distributed algorithms:

• Distributed MIP – Multiple machines work together to solve a single MIP model.
• Distributed Concurrent – Multiple machines use different algorithmic strategies to race to solve an LP or MIP model.
• Distributed Tuning – Multiple machines do experimental solves to find parameter settings that improve performance.