Intel Xeon 6 Granite Rapids Memory Scaling Performance From 6 To 12 MRDIMMs

Written by Michael Larabel in Memory on 17 February 2026 at 10:00 AM EST. Page 8 of 10. 1 Comment.
ClickHouse benchmark with settings of 100M Rows Hits Dataset, First Run / Cold Cache. 12 x MRDIMM was the fastest.
ClickHouse benchmark with settings of 100M Rows Hits Dataset, Second Run. 12 x MRDIMM was the fastest.
Apache CouchDB benchmark with settings of Bulk Size: 100, Inserts: 1000, Rounds: 30. 10 x MRDIMM was the fastest.
Apache CouchDB benchmark with settings of Bulk Size: 100, Inserts: 3000, Rounds: 30. 10 x MRDIMM was the fastest.
Memcached benchmark with settings of Set To Get Ratio: 1:5. 12 x MRDIMM was the fastest.
Memcached benchmark with settings of Set To Get Ratio: 1:10. 12 x MRDIMM was the fastest.
Memcached benchmark with settings of Set To Get Ratio: 1:100. 12 x MRDIMM was the fastest.
Pogocache benchmark with settings of Set To Get Ratio: 1:5. 10 x MRDIMM was the fastest.
Pogocache benchmark with settings of Set To Get Ratio: 1:10. 10 x MRDIMM was the fastest.
DuckDB benchmark with settings of Benchmark: Clickbench. 12 x MRDIMM was the fastest.
DuckDB benchmark with settings of Benchmark: TPC-H Parquet. 12 x MRDIMM was the fastest.
PostgreSQL benchmark with settings of Scaling Factor: 100, Clients: 800, Mode: Read Write. 12 x MRDIMM was the fastest.
PostgreSQL benchmark with settings of Scaling Factor: 1000, Clients: 800, Mode: Read Only. 12 x MRDIMM was the fastest.
PostgreSQL benchmark with settings of Scaling Factor: 1000, Clients: 800, Mode: Read Only, Average Latency. 12 x MRDIMM was the fastest.
PostgreSQL benchmark with settings of Scaling Factor: 100, Clients: 1000, Mode: Read Write. 12 x MRDIMM was the fastest.
PostgreSQL benchmark with settings of Scaling Factor: 100, Clients: 1000, Mode: Read Write, Average Latency. 12 x MRDIMM was the fastest.
PostgreSQL benchmark with settings of Scaling Factor: 1000, Clients: 1000, Mode: Read Write. 12 x MRDIMM was the fastest.
PostgreSQL benchmark with settings of Scaling Factor: 1000, Clients: 1000, Mode: Read Write, Average Latency. 12 x MRDIMM was the fastest.

For some of the database systems tested the scaling stopped after eight MRDIMMs but for others there continued to be nice performance benefits in populating all twelve memory channels with MRDIMMs.

Related Articles