Closed
Milestone
0.5.0
Most important features:
- Theodolite re-implementation with Kotlin and Quarkus
Unstarted Issues (open and unassigned)
0
Ongoing Issues (open and assigned)
0
Completed Issues (closed)
50
- Release Theodolite 0.5
- Adjust analysis notebooks for new implementation
- Enhance the configurability of patchers
- Disable Grafana login
- Adjust analysis notebooks for new implementation
- Add features of our presentation branch
- Deploy correct number of load generator instances
- Include random scheduler in Helm chart
- Finalize Masterproject
- Add documentation for the API of the SLO Server
- Rename Quarkus projekt
- Update Theodolite Job resources for Kotlin-based implementation
- Make yaml-executor mode the default
- Restructure location of new YAML files
- Add build and publish step for Lag Trend SLO Checker in CI pipeline
- Use official Theodolite container images
- Change all configurations to default values
- Add prefix to output files to link results and experiment
- Make paths configurable in TheodoliteYamlExecutor
- Add Kubernetes benchmark definitions for Kafka Streams
- Add Kubernetes benchmark definitions for Flink
- Build native images in CI pipeline
- Declare Gradle dependencies via `implementation` instead of `compile`
- Introduce Error Handling
- Restructuring to create the PatcherDefinition inside the TheodoliteExecutor
- Allow using the random scheduler in benchmarks
- Remove previous Python implementation
- Handle shutdown of Theodolite
- Use Zookeeper-less workload generator
- Make Namespace configurable or infer it from Kubernetes confguration
- Introduce the Benchmark class
- Re-implement Theodolite with Kotlin/Quarkus
- Add KDocs for all classes, interfaces, public methods
- Create a Python SLO Checker Component
- Introduce benchmark evalutation implementation in Kotlin
- Create abstract BenchmarkExecutor as base class for Benchmark executors
- Add logger
- Implement features of run_uc in Theodolite Kotlin
- Implement Search Strategies for Kotlin Prototype
- Implement Zookeeper communication with Theodolite Kotlin
- Implement Kafka communication with Theodolite Kotlin
- Add CI Pipeline for Theodolite Kotlin
- Make Kafka and Zookeeper services names configurable
- Install ServiceAccounts, Roles and RoleBindings with Helm
- Allow setting arbitrary SUT parameters from Theodolite
- Consider rewriting Theodolite in a more object-oriented way
- Add abstraction layer for stream processing engine
- Make nodeSelector configurable
- Design revised Theodolite architecture
- Introduce Theodolite Kubernetes Operator
Loading
Loading
Loading