How Amazon assigns 2 million VMs to 100,000 servers using Hexaly

Customer stories Information Technology Packing

Marion Technical Writer December 12, 2024

Rubén Ruiz, Principal Applied Scientist at Amazon and former Professor of Statistics and Operations Research at Universitat Politècnica de València, presented a talk entitled “Pragmatic OR: solving large-scale optimization problems in fast-moving environments” at the EURO Online Seminar Series.

In his seminar, Rubén Ruiz advocates OR practitioners not to focus on searching for “optimal solutions.” Instead, he advocates for “Pragmatic OR:” the delivery of OR software solutions that focus on improving the current state of the business while keeping the total cost of ownership under control. He shows how he leverages Hexaly at Amazon to solve a gigantic optimization problem that is out of the scope of MILP solvers like Gurobi and Xpress: assigning 2 million Virtual Machines (VMs) to 100,000 servers.

The replay of Rubén Ruiz’s seminar is available below.

Here is the abstract of Rubén Ruiz’s seminar.

Pragmatic OR: Solving large-scale optimization problems in fast-moving environments

This talk examines the gap between academic Operations Research and real-world industrial applications, particularly in environments like Amazon and AWS where sheer scale and delivery speed are important factors to consider. While academic research often prioritizes complex algorithms and optimal solutions, large-scale industrial problems demand more pragmatic approaches. These real-world scenarios frequently involve multiple objectives, soft constraints, and rapidly evolving business requirements that require flexibility and quick adaptation.

We argue for the use of heuristic solvers and simplified modeling techniques that prioritize speed, adaptability, and ease of implementation over strict optimality or complex approaches. This angle is particularly valuable when dealing with estimated input data, where pursuing optimality may be less meaningful.

The presentation will showcase various examples, including classical routing and scheduling problems, as well as more complex scenarios like virtual machine placement in Amazon EC2. These cases illustrate how pragmatic methods can effectively address real-world challenges, offering robust and maintainable solutions that balance performance with operational efficiency. The goal is to demonstrate that in many industrial applications, a small optimality gap is an acceptable trade-off for significantly improved flexibility and reduced operational overhead.