Repository navigation
Add Redis Cloud Pro and Trino full-data ClickBench results - #2443
Open
alberttwong wants to merge 18 commits into
Open
alberttwong wants to merge 18 commits into
alberttwong wants to merge 18 commits into
Conversation
|
|
Contributor
alberttwong
changed the title
This was referenced Oct 9, 2026
alberttwong
marked this pull request as ready for review
October 9, 2026 19:44
This branch has not been deployed
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
All 43 unchanged ClickBench queries pass independent validation on all 99,997,497 rows, three attempts each (129/129 outputs). This complete cohort uses the same merged PR154 driver d83d504; production JAR SHA256 a9e069c287a027b84be4e1bbe03ef2f6025d8e843e563e49cbde7c3d6e4e70eb. Q24 best warm: 2.991580s; Q34: 62.044117s. Original CSVs, plans, metrics, checksum manifests and validation records are retained.
The sweep initializes from a verified compatible backup and is tagged no-cold: Redis stays running while Trino restarts and runner page cache clears before each query block. Restored readiness is 545.174150s; separately measured fresh ingestion plus indexing is 1036.643924s. Restoration is setup and is never used as ingestion load_time. The one-trial restore diagnostic is 1.9015x faster. Invalid premature readiness timing is excluded. Future initialization prefers verified compatible backups.
Hardware: three r8g.16xlarge Redis hosts plus r7a.4xlarge Trino runner, AWS us-east-1 / physical AZ use1-az6, 1000 GB all RAM, 40 shards, RESP3, OSS Cluster API, Standard QPF, no replication, noeviction, AOF1s, 16000 IOPS / 1000 MiB/s data disks. Persistent data size unmeasured; no verified default API comparison. Older and targeted cohorts remain separate, with no mixed-revision timings.
Dedicated Redis Cloud subscription/database, all four EC2 instances, security group, data volumes, ephemeral artifact bucket, deadline Lambda/rule, IAM roles and instance profile are confirmed deleted. The private encrypted 40-file backup is confirmed retained with 30-day object expiry; local raw archives and validation records remain retained.
Documentation and evidence: redis-field-engineering/redis-sql-trino#156