| Age | Commit message (Collapse) | Author |
|---|
|
This change disables optimizing images for exporting, since we do not
want to depend on an external service.
|
|
This change makes the root redirect to the projects page temporary, to
prevent browsers from caching this route indefinitely, while we might
add a home page in the future.
|
|
This change removes the dependency on the Roboto font which was
downloaded for every page of OpenDC. Since we do not actually use this
font in any of our page, we can safely drop the dependency.
|
|
This change updates the web interface to use Next 13 and React 18. This
release has a couple breaking changes (related to links) which we have fixed
accordingly.
|
|
This change removes the explicit dependency on FontAwesome. PatternFly
also ships with the FontAwesome icons, so use these icons instead to
reduce the footprint of the application.
|
|
This pull request introduces the new `flow2` multi-flow simulator into the OpenDC codebase
and adjust all existing modules to make use of this new simulator.
The new simulator models flow as a network of components, which can each receive flow
from (potentially) multiple other components. In the previous simulator, the framework itself
supported only single flows between components and required re-implementation of many
components to support multiplexing flows.
Initial benchmarks show performance improvements in the range 2x–4x for large scale experiments
such as the Capelin benchmarks.
## Implementation Notes :hammer_and_pick:
* Add support for multi-flow stages
* Support flow transformations
* Add forwarding flow multiplexer
* Expose metrics on FlowMultiplexer
* Re-implement network sim using flow2
* Re-implement power sim using flow2
* Re-implement compute sim using flow2
* Optimize workload implementation of SimTrace
* Remove old flow simulator
* Add log4j-core dependency
## External Dependencies :four_leaf_clover:
* N/A
## Breaking API Changes :warning:
* Removal of the `org.opendc.simulator.flow` package. You should now use
the new flow simulator located in `org.opendc.simulator.flow2`.
* `PowerModel` interface is replaced by the `CpuPowerModel` interface.
* `PowerDriver` interface is replaced by the `SimPsu` and `SimPsuFactory` interfaces.
* Removal of `SimTraceWorkload`. Instead, create a workload from a `SimTrace` using
`createWorkload(offset)`.
* `ScalingGovernor` has been split in a `ScalingGovernor` and `ScalingGovernorFactory`.
* All modules in `opendc-simulator` are now written in Java. This means that default
parameters are not supported anymore for these modules.
|
|
This change adds the log4j-core dependency to various modules of OpenDC
using log4j2, to ensure logging keeps working. The upgrade to SLF4J 2.0 broke
the Log4j2 functionality, since the log4j-core artifact is not
automatically shipped with the SLF4J implementation.
|
|
This change removes the old version of the flow simulator from the
OpenDC repository. The old version has been replaced by the new flow2
framework which is able to simulate flows more efficiently.
|
|
This change updates the workload implementation of SimTrace by
introducing multiple implementations of the FlowStageLogic interface,
which is selected dynamically when the workload starts.
This change helps eliminate the unecessary division (and memory fetch)
when running a workload with just a single CPU.
|
|
This change re-implements the OpenDC compute simulator framework using
the new flow2 framework for modelling multi-edge flow networks. The
re-implementation is written in Java and focusses on performance and
clean API surface.
|
|
This change updates the `opendc-simulator-power` module to use the new
flow simulation framework in OpenDC (named flow2 for now).
|
|
|
|
|
|
This change implements a new `FlowMultiplexer` that forwards the inputs
directly to one of the pre-allocated outputs.
|
|
This change adds a new component, FlowTransform, which is able to
transform the flow from one component to another, based on some
inversible transformation.
Such as component is useful when converting between different units of
flow between different components.
|
|
This change adds support for creating nodes in a flow graph that support
multiple inputs and outputs directly, instead of our current approach
where we need to re-implement the `FlowConsumerContext` interface in
order to support multiple inputs or outputs.
|
|
Bumps [mikepenz/action-junit-report](https://github.com/mikepenz/action-junit-report) from 3.5.1 to 3.5.2.
- [Release notes](https://github.com/mikepenz/action-junit-report/releases)
- [Commits](https://github.com/mikepenz/action-junit-report/compare/v3.5.1...v3.5.2)
---
updated-dependencies:
- dependency-name: mikepenz/action-junit-report
dependency-type: direct:production
update-type: version-update:semver-patch
...
Signed-off-by: dependabot[bot] <support@github.com>
Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
|
|
This pull request adds support for tracking and limiting the simulation time consumed per user.
These changes allow users to consume 60 minutes of shared compute resources for simulations
in OpenDC per month.
Closes #103
## Implementation Notes :hammer_and_pick:
* Limit exception mapper to WebApplicationException
* Add support for accounting simulation time
* Add API for querying user accounting data
* Show monthly simulation budget in UI
## External Dependencies :four_leaf_clover:
* N/A
## Breaking API Changes :warning:
* The web runner now also sends the runtime of the simulation job to the server, in order to
update the consumed simulation time in the database.
|
|
This change addresses an issue with the topology sidebar where hovering
a CPU or GPU would not present the correct information due to bug.
|
|
This change fixes an issue with the OpenDC web runner where the default
job timeout was set to 10 ms instead of 10 minutes. For longer
simulations, this would cause the job to be terminated.
|
|
This change resolves an issue in the web runner where the finished VMs
would always be reported as zero.
|
|
This change updates the compute service telemetry to also expose the
number of servers that are registered with the service.
|
|
This change updates the OpenDC web UI to show the monthly simulation
budget of the user in the user dropdown. This provides the user with a
progress bar of the used simulation minutes.
|
|
This change updates the Quarkus-based web server with a new endpoint for
querying data about the active user including accounting data.
|
|
This change updates the Quarkus-based web server to add support for
tracking and limiting the simulation minutes used by the user in order
to prevent misuse of shared resources.
|
|
This change updates the Quarkus-based web server to limit the default
exception mapper to just WebApplicationException. Other exceptions
should be considered internal server errors and must not be shared with
users.
|
|
This pull request updates the build and runtime dependencies used by OpenDC to their latest
version compatible with the project.
## Implementation Notes :hammer_and_pick:
* Update simulator dependency versions
* Remove unused distribution conventions
* Update next version to 3.0
* Eliminate use of wildcard imports
* Switch to Spotless for formatting
|
|
This change updates the build configuration to use Spotless for code
formating of both Kotlin and Java.
|
|
This change updates the repository to remove the use of wildcard imports
everywhere. Wildcard imports are not allowed by default by Ktlint as
well as Google's Java style guide.
|
|
This change updates the build script in preparation for the OpenDC v3.0
release changing the version numbers to the appropriate values.
|
|
This change removes the distribution conventions used by the root
project of OpenDC. Previously, we used this to build a single
distribution for OpenDC containing the experiment uberjars and scripts
to start OpenDC.
However, with the introduction of the Quarkus-based web server, we have
decided to split releases into (potentially) multiple distributions
(e.g., one for the web server, one for just the web runner, etc.).
Furthermore, the implementation of this convention caused issues with
several other Gradle plugins.
|
|
This change updates the simulator dependencies to the latest available
version where possible.
|
|
This pull request extracts the scheduler from the `SimulationCoroutineDispatcher` into
a separate `SimulationScheduler` class which allows users to re-use the scheduler
between different coroutine dispatchers.
We implement the `SimulationScheduler` in Java, removing the explicit dependency on
Kotlin or `kotlinx-coroutines`. The scheduler uses a separate specialized priority queue
implementation that eliminates allocation in the hot path of the simulator.
## Implementation Notes :hammer_and_pick:
* Add Java-based simulator core
* Use SimulationScheduler in coroutine dispatcher
* Rename runBlockingSimulation to runSimulation
## External Dependencies :four_leaf_clover:
* N/A
## Breaking API Changes :warning:
* The Kotlin API for simulation has been moved to `org.opendc.simulator.kotlin`.
* `runBlockingSImulation` renamed to `runSimulation`
|
|
This change renames the method `runBlockingSimulation` to
`runSimulation` to put more emphasis on the simulation part of the
method. The blocking part is not that important, but this behavior is
still described in the method documentation.
|
|
This change updates the implementation of `SimulationDispatcher` to use
a (possibly user-provided) `SimulationScheduler` for managing the
execution of the simulation and future tasks.
|
|
This change introduces a new class, `SimulationScheduler`, which
provides the basis for simulations in OpenDC by allowing components to
schedule future tasks using delay-skipping queue and a virtual clock.
This new class is written in Java to remove any dependency on the
Kotlin and kotlinx-coroutines runtime when not necessary.
|
|
Bumps [mikepenz/action-junit-report](https://github.com/mikepenz/action-junit-report) from 3.5.0 to 3.5.1.
- [Release notes](https://github.com/mikepenz/action-junit-report/releases)
- [Commits](https://github.com/mikepenz/action-junit-report/compare/v3.5.0...v3.5.1)
---
updated-dependencies:
- dependency-name: mikepenz/action-junit-report
dependency-type: direct:production
update-type: version-update:semver-patch
...
Signed-off-by: dependabot[bot] <support@github.com>
Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
|
|
This pull request implements a new tool to help provision and manage the
experimental environment.
## Implementation Notes :hammer_and_pick:
* Add service registry for cloud services
* Add provisioning tool for experiments
* Add provisioning step for workflow service
* Add provisioners for FaaS service
* Use experiment base for Capelin experiments
* Use experiment base for web runner
* Integrate compute workload classes
* Remove Topology interface
## Breaking API Changes :warning:
* Removal of the `opendc-compute-workload`, `opendc-faas-workload`,
and `opendc-workflow-workload` modules. These are now located
inside `opendc-experiments`
* Removal of `ComputeServiceHelper`. Use `Provisioner` to provision
a `ComputeService`.
|
|
This change removes the Topology interface from the
`opendc-experiments-compute` module, which was meant for provisioning
the experimental topology. Howerver, with the stateless `HostSpec`
class, it is not needed to resolve the topology everytime.
|
|
This change integrates the classes from the old
`opendc-compute-workload` module into the `opendc-experiments-compute`
module. This new module contains helper classes for setting up
experiments with the OpenDC compute service.
|
|
This change updates the OpenDC web runner to use the new
`opendc-experiments-base` module for setting up the experimental
environment and simulate the workload.
|
|
This change updates the Capelin experiments to use the new
`opendc-experiments-base` module for setting up the experimental
environment and simulate the workloads.
|
|
This change adds a new module `opendc-experiments-faas` that provides
provisioner implementations for experiments to use for setting up the
FaaS service of OpenDC.
|
|
This change adds a new module `opendc-experiments-workflow` that provides
provisioner implementations for experiments to use for setting up and
using the workflow engine in OpenDC.
|
|
This change adds a new module `opendc-experiments-compute` that provides
provisioner implementations for experiments to use for setting up the
compute service of OpenDC and provisioning (simulated) hosts.
|
|
This change adds to the experiment base in OpenDC a provisioning tool to
help setup the experimental environment in a reproducible manner.
The experimental environment can be defined in terms of *provisioning
steps* which manage the allocation, configuration, and clean-up of all
resources necessary for the experiments.
The provisioning steps are executed sequentially, so in case of dependencies,
the steps need to be ordered correctly.
|
|
This change adds a new module called opendc-experiments-base which will
provide a base for doing experiments with OpenDC. The initial feature we
introduce is the service registry which acts as DNS for services to
register during experimentation.
|
|
This pull request refactors the existing workload interference model in order
to remove a dependency on it during the topology construction. With this change,
interference domains (e.g., a single host) can be constructed independently of the
interference profiles of virtual machines.
## Implementation Notes :hammer_and_pick:
* Move VM interference model into compute simulator
* Remove convergence listener parameter
* Remove FlowEngine from SimMachineContext
* Remove timestamp parameter from SimTrace
* Pass interference key via parameter
* Move interference logic into VmInterferenceMember
* Prevent boxing in interference algorithm
* Extract Random dependency from interference model
* Add separate error host state
* Simplify constructor of SimHost
* Make interference domain independent of profile
## External Dependencies :four_leaf_clover:
* N/A
## Breaking API Changes :warning:
* The interface of `VmInterferenceModel` is changed. Users do not need to
provide a seed for the model anymore.
* A `VmInterferenceModel` should be passed via the metadata parameter of
`startWorkload` to enable workload interference.
|
|
This change updates the interface of `ComputeService` to provide access
to the instances (servers) that have been registered with the compute
service. This allows metric collectors to query the metrics of the
servers that are currently running.
|
|
This change updates the `ComputeServiceHelper` class to provide the
failure model via a parameter to the `run` method instead of constructor
parameter. This separates the construction of the topology from the
simulation of the workload.
|
