You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Some performance testing of the bulk import approaches has yielded the following results. 10B randomly generated records, with a simple schema (key:string, timestamp:int64, value:string), were ingested from 160 Parquet files with zstd compression, with a total size of ~175.5 GB.
The tests show that Graviton based m7g instances are faster and cheaper than the current default instance types (m7i).
The text was updated successfully, but these errors were encountered:
Some performance testing of the bulk import approaches has yielded the following results. 10B randomly generated records, with a simple schema (key:string, timestamp:int64, value:string), were ingested from 160 Parquet files with zstd compression, with a total size of ~175.5 GB.
The tests show that Graviton based m7g instances are faster and cheaper than the current default instance types (m7i).
The text was updated successfully, but these errors were encountered: