Question with using METIS in AMG

[txthpc]

Hello, I have a question about aggregation using Metis in AMG (algebraic multigrid) and want to discuss it with you
Due to the limited number of split subgraphs imposed by metis, when the fine grid matrix is large enough, Metis cannot aggregate enough coarse nodes (for example, there are one million nodes in the fine grid layer, where each two nodes aggregate into points on the next grid layer, which will result in about 500,000 aggregations, and Metis cannot split a graph of this size).
Do you have any good suggestions for this situation?
I look forward to receiving your reply, and I will benefit from any help you give me.

[karypis]

Have you looked at http://glaros.dtc.umn.edu/gkhome/mgridgen/overview ?

…

________________________________ From: txthpc ***@***.***> Sent: Thursday, May 12, 2022 12:45:33 AM To: KarypisLab/METIS ***@***.***> Cc: Subscribed ***@***.***> Subject: [KarypisLab/METIS] Question with using METIS in AMG (Issue #30) Hello, I have a question about aggregation using Metis in AMG (algebraic multigrid) and want to discuss it with you Due to the limited number of split subgraphs imposed by metis, when the fine grid matrix is large enough, Metis cannot aggregate enough coarse nodes (for example, there are one million nodes in the fine grid layer, where each two nodes aggregate into points on the next grid layer, which will result in about 500,000 aggregations, and Metis cannot split a graph of this size). Do you have any good suggestions for this situation? I look forward to receiving your reply, and I will benefit from any help you give me. — Reply to this email directly, view it on GitHub<#30>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/AAI63B54OCML7KXY6NC3JILVJSZJ3ANCNFSM5VXKD6GA>. You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>