Import

August 22, 2026 · View on GitHub

Systems

SystemProviderCPUCoresRAMSSD
LEA25on-premEPYC 7543P432 GiB3.2 TB Intel P5600 over vSAN
LEA36on-premEPYC 7543P864 GiB3.2 TB Intel P5600 over vSAN
LEA47on-premEPYC 7543P16128 GiB3.2 TB Intel P5600 over vSAN
LEA58on-premEPYC 7543P32256 GiB3.2 TB Intel P5600 over vSAN
LEA79on-premEPYC 9555128768 GiB12.8 TB Huawei OceanDisk 300P
A5N46on-premRyzen 9900X2496 GiB4 TB Samsung 990 Pro

All systems were configured according to the Production Configuration guide.

On all systems, the heap memory and the block cache were each configured to use 1/4 of the total available RAM. Consequently, the Blaze process itself uses about half of the available system memory, leaving the remainder for the file system cache.

Datasets

The following datasets were used:

DatasetHistory# Pat. ¹# Res. ²# Obs. ³Size on SSD
100k10 years100 k104 M59 M110 GiB
100k-fhfull100 k317 M191 M323 GiB
1M10 years1 M1044 M593 M1045 GiB

¹ Number of Patients, ² Total Number of Resources, ³ Number of Observations

The creation of the datasets is described in the Synthea section. The disc size is measured after full manual compaction of the database. The actual disc size will be up to 50% higher, depending on the state of compaction which happens regularly in the background.

Import

blazectl --server http://localhost:8080/fhir upload -c8 <dataset>
SystemDatasetDB Scale FactorMax SubcompactionsTime (h)Resources/s
LEA47100k112.85510108
LEA47100k-fh119.7978977
LEA58100k112.16913305
LEA58100k-fh117.42611843
LEA581M4126,48010955
LEA79100k810,64844520
LEA79100k880,42667737
LEA791M1118.22715915
LEA791M4112.17423829

Read more about the DB Scale Factor and Max Subcompactions in the Environment Variables Section.