v2.0.0-RC2

What's Changed

Listing all changes since the version used for SNB Interactive v1 audits (0.3.6):

• Include empty results in validation output and use them for validation by @szarnyasg in #146
• Rename packages from 'com.ldbc.*' to 'org.ldbcouncil.*' by @szarnyasg in #147
• Include Scale Factors in driver by @GLaDAP in #148
• Feature/fix tests by @GLaDAP in #149
• Use more verbose logging in CI by @szarnyasg in #150
• Add script to package Maven artifacts by @szarnyasg in #151
• Feature/fix validation mode by @GLaDAP in #152
• Feature/separate validation properties by @GLaDAP in #155
  • Separates the validation properties. New validation properties are introduced.
  • Introduce the mode property to make OperationMode choice explicit. This replaces the create_validation_parameters and create_workload_statistics properties (fixes #120)
  • Fix validation to be compliant with specification: the 95% on-time requirement for passing a "schedule audit" is now implemented correctly. Previously (in v0.x) it was a constant amount of 10 operations per operation type (fixes #154)
  • Parameter parameters_dir is now skipped during database validation mode (fixes #144)
• Add missing newlines by @szarnyasg in #156
• Fix serialization validation parameters. Cleanup Workload class by @GLaDAP in #157
• Change replacement parameters in read queries. Remove unused class by @szarnyasg in #158
• Fix validation errors by @GLaDAP in #159
• Merge update streams by @GLaDAP in #164
• Feature/fix query 14 validation by @GLaDAP in #177
• Update sbe-tools to agrona by @GLaDAP in #180
• Batch update stream events by @GLaDAP in #179
• Upgrade junit and Hamcrest by @GLaDAP in #181
• Feature/fix conversion scripts by @GLaDAP in #183
• Rename path for conversion script by @GLaDAP in #184
• Rename path variable by @GLaDAP in #185
• Add remove_suffix function for Python <3.9 compatibility by @GLaDAP in #186
• Output dependent-time added parquet files in update-stream folder by @GLaDAP in #187
• Fix insert validation parameter generation by @GLaDAP in #190
• Check if operation stream has next by @GLaDAP in #191
• Parameter curation for v2.0 by @szarnyasg in #189
• Asynchronous update streams by @GLaDAP in #192
• Refactor tests by @GLaDAP in #193
• Validation: check 5% delay per query type instead of 5% of total operation count by @GLaDAP in #194
• Use PersonKnowsPersonDays factor table for path selection by @GLaDAP in #196
• Feature/operation scheduling fix by @GLaDAP in #200
• Feature/fix path parameter curation by @GLaDAP in #201
• Optimize conversion scripts by @szarnyasg in #202
• Optimize creation of update streams with dependencyTime column by @GLaDAP in #203
• Make batch size configurable and auto-detect Person_hasInterestTag.{TagId,interestId} attribute name by @GLaDAP in #205
• Add Forum_hasMember_Person to batched update stream creation by @GLaDAP in #207
• Merge consecutive dates. Add arg option for path finding script by @GLaDAP in #209
• Fix datetime columns in paramgen by @GLaDAP in #211
• Fix scheduling of dependent operations by @GLaDAP in #213
• Conversion script: Batch Person_likes_{Comment,Post}, adjust batch size by @szarnyasg in #208
• Merge parameters on consecutive days by @GLaDAP in #216
• Data conversion script: Add more logging to the create_update_streams.py script by @szarnyasg in #217
• CI: Test conversion script and paramgen on the SF1 data set by @szarnyasg in #218
• Allow for smaller batch sizes by @GLaDAP in #221
• Conversion script: Bump DuckDB version to 0.6.1 by @szarnyasg in #220
• Conversion script: Use environment variable 'LDBC_SNB_DATA_ROOT_DIRECTORY' consistently by @szarnyasg in #222
• Paramgen enforce factor dir arg by @szarnyasg in #223
• Update parameter generation by @GLaDAP in #224
• Move temporal parameter check to merging iterator by @GLaDAP in #225
• Paramgen: dynamically determine paramgen thresholds by @GLaDAP in #226
• Feature/upgrade to java 17 by @GLaDAP in #227
• Bump DuckDB version to 0.7.1 by @szarnyasg in #228

Full Changelog: 0.3.6...v2.0.0-RC2

v2.0.0-RC1

What's Changed

• Feature/fix query 14 validation by @GLaDAP in #177
• Update sbe-tools to agrona by @GLaDAP in #180
• Batch update stream events by @GLaDAP in #179
• Upgrade junit and Hamcrest by @GLaDAP in #181
• Feature/fix conversion scripts by @GLaDAP in #183
• Rename path for conversion script by @GLaDAP in #184
• Rename path variable by @GLaDAP in #185
• Add remove_suffix function for Python <3.9 compatibility by @GLaDAP in #186
• Output dependent-time added parquet files in update-stream folder by @GLaDAP in #187
• Fix insert validation parameter generation by @GLaDAP in #190
• Check if operation stream has next by @GLaDAP in #191
• Parameter curation for v2.0 by @szarnyasg in #189
• Asynchronous update streams by @GLaDAP in #192
• Refactor tests by @GLaDAP in #193
• Validation: check 5% delay per query type instead of 5% of total operation count by @GLaDAP in #194
• Use PersonKnowsPersonDays factor table for path selection by @GLaDAP in #196
• Feature/operation scheduling fix by @GLaDAP in #200
• Feature/fix path parameter curation by @GLaDAP in #201
• Optimize conversion scripts by @szarnyasg in #202
• Optimize creation of update streams with dependencyTime column by @GLaDAP in #203
• Make batch size configurable and auto-detect Person_hasInterestTag.{TagId,interestId} attribute name by @GLaDAP in #205
• Add Forum_hasMember_Person to batched update stream creation by @GLaDAP in #207
• Merge consecutive dates. Add arg option for path finding script by @GLaDAP in #209
• Fix datetime columns in paramgen by @GLaDAP in #211
• Fix scheduling of dependent operations by @GLaDAP in #213
• Conversion script: Batch Person_likes_{Comment,Post}, adjust batch size by @szarnyasg in #208
• Merge parameters on consecutive days by @GLaDAP in #216
• Data conversion script: Add more logging to the create_update_streams.py script by @szarnyasg in #217
• CI: Test conversion script and paramgen on the SF1 data set by @szarnyasg in #218
• Allow for smaller batch sizes by @GLaDAP in #221

Full Changelog: v2.0.0...v2.0.0-RC1

v1.2.0

Contains the improvements of v1.0.0 but reverts PR #164 that merged update streams. This caused increased memory consumption and introduced performance problems.

v2.0.0

What's Changed

• Feature/fix operation stream loading by @GLaDAP in #165
• Feature/fix validation params by @GLaDAP in #170
• Use parquet loader by @szarnyasg in #172
• Change counting of validation parameter creation by @GLaDAP in #174
• Rename LdbcUpdate* classes to LdbcInsert* by @szarnyasg in #175
• v2.0 with deletes by @szarnyasg in #169

Full Changelog: v1.0.0...v2.0.0

v1.0.0

Initial release after the refactoring.

What's Changed

• Include empty results in validation output and use them for validation by @szarnyasg in #146
• Rename packages from 'com.ldbc.*' to 'org.ldbcouncil.*' by @szarnyasg in #147
• Include Scale Factors in driver by @GLaDAP in #148
• Feature/fix tests by @GLaDAP in #149
• Use more verbose logging in CI by @szarnyasg in #150
• Add script to package Maven artifacts by @szarnyasg in #151
• Feature/fix validation mode by @GLaDAP in #152
• Feature/separate validation properties by @GLaDAP in #155
  • Separates the validation properties. New validation properties are introduced.
  • Introduce the mode property to make OperationMode choice explicit. This replaces the create_validation_parameters and create_workload_statistics properties (fixes #120)
  • Fix validation to be compliant with specification: the 95% on-time requirement for passing a "schedule audit" is now implemented correctly. Previously (in v0.x) it was a constant amount of 10 operations per operation type (fixes #154)
  • Parameter parameters_dir is now skipped during database validation mode (fixes #144)
• Add missing newlines by @szarnyasg in #156
• Fix serialization validation parameters. Cleanup Workload class by @GLaDAP in #157
• Change replacement parameters in read queries. Remove unused class by @szarnyasg in #158
• Fix validation errors by @GLaDAP in #159
• Merge update streams by @GLaDAP in #164

New Contributors

• @GLaDAP made their first contribution in #148

Full Changelog: 0.3.6...v1.0.0

0.3.6

Only documentation improvements and cleanup.

0.3.5

Changes:

• Removed the BI classes. (The BI workload will be implemented as a Python driver.)

0.3.4

Change:

• Add -flush_log option to flush log after each operation (useful for ACID/durability tests)

0.3.3

This release only contains cosmetic changes compared to v0.3.2.

0.3.2

Merge pull request #105 from ldbc/dev

Release 0.3.2