Memory overflow after updating to a new version.

Answered

abc abc

August 07, 2023 10:34 Edited

• 
• 

Dear Gurobi Support Team

On a 32-core server with 64 threads, I solve a MILP model in each core using 2 threads, so there are 32 models being solved in parallel. The problem is, when using Gurobi 9.1.2, all these models can be solved at a cost of about 100GB memory. But after updating to Gurobi 9.5.2 or 10.0.2, the memory consumption increases sharply and soon it exceeds 200 GB, then the program runs into a out-of-memory error. There is no parameter setting in each MILP model (using default setting). So I was wondering what makes the memory consumption so different? and Is there any way to solve this problem?

Thanks a lot in advance

Related to

• Memory

0

• 

  David Torres Sanchez

  • Gurobi Staff

  August 07, 2023 11:55

  Hi,

  We have turned this request into an internal ticket.

  Cheers, 
  David

  0

Please sign in to leave a comment.