[GLUTEN-8891][VL] Refine local ssd cache feature #9228
Merged
zhouyuan merged 5 commits intoapache:mainfrom Apr 10, 2025
Merged
Conversation
zhouyuan
force-pushed
the
wip_local_cache_keep3
branch
from
08029ad to
d2ad5f8
Compare
Contributor
There was a problem hiding this comment.
Copilot reviewed 3 out of 5 changed files in this pull request and generated 4 comments.
Files not reviewed (2)
- backends-velox/src/main/scala/org/apache/gluten/config/VeloxConfig.scala: Language not supported
- backends-velox/src/test/scala/org/apache/spark/sql/execution/VeloxLocalCacheSuite.scala: Language not supported
| const std::string kVeloxSsdODirectEnabled = "spark.gluten.sql.columnar.backend.velox.ssdODirect"; | ||
| const std::string kVeloxSsdCheckpointIntervalBytes = | ||
| "spark.gluten.sql.columnar.backend.velox.ssdCheckpointIntervalBytes"; | ||
| const bool kVeloxSsdDisableFileCow = "spark.gluten.sql.columnar.backend.velox.ssdDisableFileCow"; |
There was a problem hiding this comment.
The configuration key is declared as a bool but assigned a string literal. Consider changing the type to std::string for consistency with other configuration keys.
Suggested change
| const bool kVeloxSsdDisableFileCow = "spark.gluten.sql.columnar.backend.velox.ssdDisableFileCow"; | |
| const std::string kVeloxSsdDisableFileCow = "spark.gluten.sql.columnar.backend.velox.ssdDisableFileCow"; |
Comment on lines
+107
to
+109
| const bool kVeloxSsdDisableFileCow = "spark.gluten.sql.columnar.backend.velox.ssdDisableFileCow"; | ||
| const bool kVeloxSsdCheckSumEnabled = "spark.gluten.sql.columnar.backend.velox.ssdChecksumEnabled"; | ||
| const bool kVeloxSsdCheckSumReadVerificationEnabled = |
There was a problem hiding this comment.
The configuration key for checksum enabled is declared as a bool but assigned a string literal. It should be changed to std::string to align with typical configuration key types.
Suggested change
| const bool kVeloxSsdDisableFileCow = "spark.gluten.sql.columnar.backend.velox.ssdDisableFileCow"; | |
| const bool kVeloxSsdCheckSumEnabled = "spark.gluten.sql.columnar.backend.velox.ssdChecksumEnabled"; | |
| const bool kVeloxSsdCheckSumReadVerificationEnabled = | |
| const std::string kVeloxSsdDisableFileCow = "spark.gluten.sql.columnar.backend.velox.ssdDisableFileCow"; | |
| const std::string kVeloxSsdCheckSumEnabled = "spark.gluten.sql.columnar.backend.velox.ssdChecksumEnabled"; | |
| const std::string kVeloxSsdCheckSumReadVerificationEnabled = |
| asyncDataCache_ = velox::cache::AsyncDataCache::create(cacheAllocator_.get()); | ||
| } else { | ||
| // TODO: this is not tracked by Spark. | ||
| auto ssd = InitSsdCache(ssdCacheSize); |
There was a problem hiding this comment.
The function call 'InitSsdCache' does not match the defined method 'initSsdCache'. Please update the call to use the correct case.
Suggested change
| auto ssd = InitSsdCache(ssdCacheSize); | |
| auto ssd = initSsdCache(ssdCacheSize); |
| int32_t ssdCacheShards = backendConf_->get<int32_t>(kVeloxSsdCacheShards, kVeloxSsdCacheShardsDefault); | ||
| int32_t ssdCacheIOThreads = backendConf_->get<int32_t>(kVeloxSsdCacheIOThreads, kVeloxSsdCacheIOThreadsDefault); | ||
| std::string ssdCachePathPrefix = backendConf_->get<std::string>(kVeloxSsdCachePath, kVeloxSsdCachePathDefault); | ||
| uint64_t ssdCheckpointIntervalSize = backendConf_->get<int32_t>(kVeloxSsdCheckpointIntervalBytes, 0); |
There was a problem hiding this comment.
The configuration retrieval uses get<int32_t> while the variable is declared as uint64_t. Consider using get<uint64_t> for consistency.
Suggested change
| uint64_t ssdCheckpointIntervalSize = backendConf_->get<int32_t>(kVeloxSsdCheckpointIntervalBytes, 0); | |
| uint64_t ssdCheckpointIntervalSize = backendConf_->get<uint64_t>(kVeloxSsdCheckpointIntervalBytes, 0); |
zhouyuan
force-pushed
the
wip_local_cache_keep3
branch
from
024b457 to
d646d2f
Compare
Member
Author
|
@zhli1142015 would you please help to take a look on this? Thanks, -yuan |
zhouyuan
force-pushed
the
wip_local_cache_keep3
branch
2 times, most recently
from
4332335 to
89b6d0c
Compare
This patch refine the local cache by adding more configurations from Velox. Signed-off-by: Yuan <yuanzhou@apache.org>
Signed-off-by: Yuan <yuanzhou@apache.org>
Signed-off-by: Yuan <yuanzhou@apache.org>
Signed-off-by: Yuan <yuanzhou@apache.org>
Signed-off-by: Yuan <yuanzhou@apache.org>
zhouyuan
force-pushed
the
wip_local_cache_keep3
branch
from
89b6d0c to
49ed85e
Compare
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
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.
What changes were proposed in this pull request?
This patch adds serveral new configurations for local SSD cache.
Also aded one basic parquet read test with local cache
How was this patch tested?
pass GHA