Hyper: load native benchmark data from Parquet

[caetanosauer]

Hyper: load native benchmark data from Parquet

Switch the native Hyper benchmark from the 81 GB CSV input to the 15 GB
single-file Parquet dataset. This reduces the measured load time from 327.0
seconds to 187.1 seconds (43%) while producing the same 99,997,497 rows and
the same logical results for all 43 ClickBench queries.

[github-actions]

Results for hyper are ready for: c6a.4xlarge.
The result files are committed as 22f613d.

Logs:

• hyper on c6a.4xlarge: benchmark log, cloud-init log

[github-actions]

The run of hyper on c8g.4xlarge did not produce results.
The run of hyper on c8g.metal-48xl did not produce results.

Logs:

• hyper on c8g.4xlarge: benchmark log, cloud-init log
• hyper on c8g.metal-48xl: benchmark log, cloud-init log

[github-actions]

Results for hyper are ready for: c6a.2xlarge, c6a.4xlarge, c6a.metal, c7a.metal-48xl.
The result files are committed as 9bdd3fb.
The run of hyper on t3a.small did not produce results.

Logs:

• hyper on t3a.small: benchmark log, cloud-init log
• hyper on c7a.metal-48xl: benchmark log, cloud-init log
• hyper on c6a.metal: benchmark log, cloud-init log
• hyper on c6a.4xlarge: benchmark log, cloud-init log
• hyper on c6a.2xlarge: benchmark log, cloud-init log

[github-actions]

Results for hyper are ready for: c6a.xlarge.
The result files are committed as ac27e70.

Logs:

• hyper on c6a.xlarge: benchmark log, cloud-init log

[github-actions]

Results for hyper are ready for: c6a.large.
The result files are committed as e755cd9.

Logs:

• hyper on c6a.large: benchmark log, cloud-init log

[vogelsgesang]

Are those results actually measured against this PR rebased on top of latest main? Do they also include #1376 already? Or do we need to rebase the PR?

[alexey-milovidov]

I remember I resolved a conflict and merged with master here: e235fcc

Likely, they reflect the latest changes. If not, we can rerun.