| Age | Commit message (Collapse) | Author |
|---|
|
* Fixed a small bug which caused the PowerSource to not always report the correct energy usage
* spotless
|
|
|
|
* Updated the FlowDistributor to work in more cases and be more performant.
* Removed old FlowDistributor
|
|
cannot be scheduled twice. (#284)
Updated the FlowNodeQueue
|
|
state. (#283)
|
|
* Restructured opendc-simulator-flow.
Renamed Multiplexer to FlowDistributor.
* spotless applied
* Added FlowDistributor topologies back
|
|
|
|
Multiplexer is using MaxMinFairness given that this is currently the default and only fairness available in OpenDC. (#280)
|
|
* Fixed the Multiplexer.java to properly divide the supply over the different consumers.
Fixed a bug where fragments were being loaded in reversed order.
* Optimized the Multiplexer.java, by only updating the supply of the consumer that updated its demand when possible.
|
|
|
|
consumerIndex variable to FlowEdge.java (#275)
|
|
* Connected the cpu power model given by the topology.json to SimCpu.java
* ran spotless
|
|
|
|
* Fixed a small bug making the power source not update energy usage properly
* small update to the test
|
|
* Updated a lot of documentation, added a new get-started tutorial.
* Applied Spotless
* Applied Spotless Java
* Added bitbrains workload to site
|
|
|
|
* Added power sources to OpenDC.
In the current form each Cluster has a single power source that is connected to all hosts in that cluster
* Added power sources to OpenDC.
In the current form each Cluster has a single power source that is connected to all hosts in that cluster
* Ran spotless Kotlin and Java
|
|
* Updated tests
Changed all floats into doubles to have consistency over the whole framework
Made a small update to the multiplexer to better push through supply and demand
Fixed small typo
Updated M3SA paths.
fixed merge conflicts
Removed unused components. Updated tests.
Improved checkpointing model
Improved model, started with SimPowerSource
implemented FailureModels and Checkpointing
First working version
midway commit
first update
All simulation are now run with a single CPU and single MemoryUnit. multi CPUs are combined into one. This is for performance and explainability.
* Updated test memory
|
|
* Removed unused components. Updated tests.
Improved checkpointing model
Improved model, started with SimPowerSource
implemented FailureModels and Checkpointing
First working version
midway commit
first update
All simulation are now run with a single CPU and single MemoryUnit. multi CPUs are combined into one. This is for performance and explainability.
* fixed merge conflicts
* Updated M3SA paths.
* Fixed small typo
|
|
CPUs are combined into one. This is for performance and explainability. (#255)
|
|
* Updated the checkpointing system to use SimTrace. The checkpoint model can now also scale, which means the interval between checkpoints can increase or decrease over time.
* spotless kotlin
* Fixed tests
* spotless apply
|
|
* Started on reimplementing the SimTrace implementation
* updated trace format. Fragments now do not have a deadline, but a duration. The Fragments are executed in order.
|
|
* Updated SimTrace to use a single ArrayDeque instead of three separate lists for deadline, cpuUsage, and coreCount
* Renamed input files to tasks.parquet and fragments.parquet. Renamed server to task. OpenDC nows exports tasks.parquet instead of server.parquet
|
|
for deadline, cpuUsage, and coreCount (#245)
|
|
|
|
* Fixed a problem which caused the CPU limit to be much lower than it should be.
AllocationPolicy is now properly exposed to the user
* Fixed tests
* spotless kotlin
|
|
|
|
* Updated the power models and added tests
* Updated test topologies
|
|
* sync with the master branch
* rebase
* multimodel - simulation is currently run as many times as you can see a model
* factory method - handles models without given params
* removed redundant flags
* modelType
* flags removed
* implemented output into a folder
* multimodel ipynb setup - to be implemented and also ran as a python script, when the simulation occurs
* towards a mutimodel python implementation - issue observed - the saved files have same data?
* json parsing handles now lists for topology, workloads, allocaitonPolicies, powerModels
* scenarioFile inputs lists, and creates multiple combinations of scenarios
* multi-model prediction repaired, now we predict using multiple models
* commit before removing powerModel from scenario
* commit after removing powerModel from scenario
* commit after removing powerModel from scenario (and actually running)
* powermodels now can output their name and full name (with min and max)
* now we can select where to output (seed or output folder)
* input files - clear naming + output naming improved
* minimal changes
* all tests passing + json files from tests updated to the new json format
* json files from topology now accept only one power model (instead of list)
* json files from topology now accept only one power model (instead of list)
* multi and single input from tests updated to match the format
* tests passed locally
* spotless applies
* demo folder removed
|
|
* Updated the topology format to JSON. Updated TopologyReader.kt to handle JSON filed. Added documentation for the new format.
* applied spotless kotlin
* small update
* Updated for spotless apply
* Updated for spotless apply
|
|
* Updated all package versions including kotlin. Updated all web-server tests to run.
* Changed the java version of the tests. OpenDC now only supports java 19.
* small update
* test update
* new update
* updated docker version to 19
* updated docker version to 19
|
|
* Updated metrics and parquet output
* fixed typos
|
|
* removed experiment-compute and integrated all components into opendc-compute
* updated workflow gradle file
* removed unneeded code
|
|
* made sure all tests run
* fixed typo
* executed spotlessApply
* added back web-server tests
* updated SimTraceWorkloadTest
* commented CapelinRunneer and GreenifierRunner tests
* commented one SimTraceWorkloadTest
* altered codecov execution
* changed codecov
|
|
|
|
|
|
This change fixes an issue with the `ForwardingFlowMultiplexer` where
the capacity of new outlets were not recorded correctly due to no
handler being attached to idle outlets, causing the `pull` events to be
disregarded.
This bug manifested in an issue where the CPU counters where reporting
negative values. This was caused by the CPU usage/demand being
subtracted from a zero capacity.
|
|
This change removes the use of a Spy object from the TestInlet class.
Since the Spy is not actually used in our tests, we remove it due to the
high initialization cost.
|
|
This change replaces the use of `CoroutineContext` for passing the
`SimulationDispatcher` across the different modules of OpenDC by the
lightweight `Dispatcher` interface of the OpenDC common module.
|
|
This change updates the `SimulationScheduler` class to implement the
`Dispatcher` interface from the OpenDC Common module, so that OpenDC
modules only need to depend on the common module for dispatching future
task (possibly in simulation).
|
|
This change updates the modules of OpenDC to always accept
the `InstantSource` interface as source of time. Previously we used
`java.time.Clock`, but this class is bound to a time zone which does not
make sense for our use-cases.
Since `java.time.Clock` implements `java.time.InstantSource`, it can be
used in places that require an `InstantSource` as parameter. Conversion
from `InstantSource` to `Clock` is also possible by invoking
`InstantSource#withZone`.
|
|
This change updates the interface of `SimWorkload` to support
snapshotting workloads. We introduce a new method `snapshot()` to this
interface which returns a new `SimWorkload` that can be started at a
later point in time and on another `SimMachine`, which continues
progress from the moment the workload was snapshotted.
|
|
This change updates the implementation of `SimMachineContext` to report
exceptions thrown in `onStop` as suppressed exceptions if an exception
caused the workload to stop.
|
|
This change adds a new method to `FlowStage` called `sync()` which
enables users to synchronously update the stage at the current
timestamp.
This functionality is neccessary to support snapshotting since we need
to synchronize the state of the `FlowStage` before creating a snapshot
of the workload.
|
|
This change adds a new static method `chain` to `SimWorkloads` to chain
multiple workloads sequentially.
|
|
This change updates the interface of `SimMachineContext` to allow
workloads to reset all resources provided by the machine to the
workload. This allows us to implement a `SimWorkload` that can compose
multiple workloads.
|
|
This change introduces a new class SimWorkloads which provides
construction methods for the standard workloads available in OpenDC.
|
|
This change updates the interface of `SimMachine#startWorkload` to
introduce a parameter `completion` that is invoked when the workload
completes either succesfully or due to failure.
This functionality has often been implemented by wrapping a
`SimWorkload` and catching its exceptions. However, since this
functionality is used in all usages of `SimMachine#startWorkload` we
instead embed it into `SimMachine` itself.
|
|
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.
|
