awslabs/benchmark-ai issues and pull requests

#1055 - nit change

Pull Request - State: open - Opened by tejaschumbalkar over 2 years ago

#1054 - Update version of terraform and EKS cluster

Pull Request - State: closed - Opened by Chancebair over 4 years ago - 1 comment

#1053 - Updating Versions

Pull Request - State: closed - Opened by Chancebair over 4 years ago

#1052 - Anubis-setup outputs var

Issue - State: closed - Opened by ryansteakley over 4 years ago - 1 comment

#1051 - Misc Housekeeping and bug fixing

Pull Request - State: open - Opened by gavinmbell over 4 years ago - 1 comment

#1050 - [sm-metrics] SM metrics using descriptor output.metrics

Pull Request - State: closed - Opened by ryansteakley over 4 years ago

#1049 - [feature] Automatic dashboard creation

Pull Request - State: closed - Opened by ryansteakley over 4 years ago - 1 comment

#1048 - [sm-executor][feature] Merge metrics

Pull Request - State: closed - Opened by ryansteakley over 4 years ago

#1047 - [feature] Sagemaker training job metrics

Pull Request - State: closed - Opened by ryansteakley over 4 years ago - 2 comments

#1046 - Documentation Cleanup

Pull Request - State: closed - Opened by gavinmbell over 4 years ago

#1045 - removing duplicates

Pull Request - State: closed - Opened by tejaschumbalkar over 4 years ago

#1044 - [feature] Custom-parameters (hyperparameters)

Pull Request - State: closed - Opened by ryansteakley over 4 years ago

#1043 - Implementing suggested changes from #1041

Issue - State: closed - Opened by tejaschumbalkar over 4 years ago - 1 comment

#1042 - [SM-EXECUTOR][FIX] Pin SageMaker to < v2.0

Pull Request - State: closed - Opened by ryansteakley over 4 years ago

#1041 - Mpijob migration

Pull Request - State: closed - Opened by tejaschumbalkar over 4 years ago - 3 comments

#1040 - Reverting #974

Pull Request - State: closed - Opened by tejaschumbalkar over 4 years ago - 2 comments

#1039 - fixing logging TypeError exception

Pull Request - State: closed - Opened by tejaschumbalkar over 4 years ago - 1 comment

#1038 - Removed ksonnet. Use kubeflow mpi-operator instead.

Pull Request - State: closed - Opened by tejaschumbalkar over 4 years ago

#1037 - Test failures for job-status-trigger and sm-executor modules

Issue - State: closed - Opened by tejaschumbalkar over 4 years ago - 2 comments

#1036 - Upgrade MPI operator and k8s config templates

Pull Request - State: closed - Opened by tejaschumbalkar over 4 years ago - 4 comments

#1035 - Is there a definitive FAIL / SUCCEED event status?

Issue - State: open - Opened by gavinmbell over 4 years ago - 1 comment
Labels: enhancement, good first issue, bai-client

#1034 - Removed ksonnet. Use kubeflow mpi-operator instead.

Pull Request - State: closed - Opened by SergTogul over 4 years ago - 5 comments

#1033 - [feature] Add --terminate command to allow for smart-exiting of the watcher

Pull Request - State: closed - Opened by ryansteakley over 4 years ago - 1 comment

#1032 - [Feature] Get/Set client-id through commandline

Pull Request - State: closed - Opened by ryansteakley over 4 years ago - 3 comments

#1031 - [feature] Add command line option to explicitly whitelist CIDR blocks

Pull Request - State: closed - Opened by ryansteakley over 4 years ago - 3 comments

#1030 - Performance drop on single node cpu run - vanilla EC2 vs Anubis

Issue - State: open - Opened by surajkota over 4 years ago
Labels: p0

#1029 - Fix Flake8 update and pin version

Pull Request - State: closed - Opened by Chancebair over 4 years ago - 2 comments

#1028 - Fix Flake8 update and pin version

Pull Request - State: closed - Opened by Chancebair over 4 years ago - 1 comment

#1027 - [Improvement] Remove --ignore=E902 in python-common.mk

Issue - State: open - Opened by Chancebair over 4 years ago
Labels: good first issue, clean-up

#1026 - [Bug] Lint Failures on Build Stage

Issue - State: closed - Opened by Chancebair over 4 years ago
Labels: bug

#1025 - Custom toml file for EKS performance test for TF is not running

Issue - State: open - Opened by TusharKanekiDey over 4 years ago - 2 comments
Labels: p0

#1024 - [Bug] [Baictl] Codepipeline Source Verification step

Issue - State: open - Opened by Chancebair over 4 years ago
Labels: bug

#1023 - In Makefile use PROJECT in all places where feasible

Pull Request - State: closed - Opened by gavinmbell almost 5 years ago - 1 comment

#1022 - [Install] [Bug] ksonnet install broken

Issue - State: open - Opened by Chancebair almost 5 years ago - 1 comment
Labels: bug, baictl, infrastructure, p0

#1021 - fix: missing script-mode functionality

Pull Request - State: closed - Opened by surajkota about 5 years ago

#1020 - change: make task_name mandatory for labels

Pull Request - State: closed - Opened by surajkota about 5 years ago - 1 comment

#1019 - fix format bugs

Pull Request - State: closed - Opened by YangFei1990 about 5 years ago

#1018 - Add FSx into the test yamls

Pull Request - State: closed - Opened by YangFei1990 about 5 years ago

#1017 - Add FSx support for yaml template

Pull Request - State: closed - Opened by YangFei1990 about 5 years ago

#1016 - [Executor] script mode not working

Issue - State: open - Opened by surajkota about 5 years ago - 1 comment
Labels: service:executor, p0

#1015 - Plotting graphs for cron jobs

Issue - State: open - Opened by surajkota about 5 years ago - 5 comments
Labels: bug, p0, RM:user-metrics, RM:AWS-CloudWatchExport

#1014 - Use FSx for data in Anubis

Issue - State: open - Opened by haohanchen-yagao about 5 years ago
Labels: enhancement

#1013 - Make anubis endpoint reachable from Native AWS

Issue - State: closed - Opened by surajkota about 5 years ago - 8 comments
Labels: bug, infrastructure, blocked, p0

#1012 - [SM-Executor] Adds support for 'args' for SageMaker benchmarks

Pull Request - State: closed - Opened by perdasilva about 5 years ago - 3 comments

#1011 - Race condition for iam perms vs container initialization.

Issue - State: open - Opened by gavinmbell about 5 years ago
Labels: p2

#1010 - Set maximum of 8 custom labels in descriptor schema

Pull Request - State: closed - Opened by jlcontreras about 5 years ago

#1009 - Limit size of info.labels to 8

Issue - State: closed - Opened by jlcontreras about 5 years ago
Labels: bai-bff

#1008 - Fix metrics-pusher support for custom labels

Pull Request - State: closed - Opened by jlcontreras about 5 years ago

#1007 - Use toml literal strings to declare metric patterns

Pull Request - State: closed - Opened by jlcontreras about 5 years ago

#1006 - [Executor] Re-enable script mode integration test

Issue - State: open - Opened by perdasilva about 5 years ago
Labels: bug, service:executor

#1005 - Using IAM roles instead or IAM users for setting up and accessing anubis

Issue - State: open - Opened by surajkota about 5 years ago
Labels: infrastructure, documentation, p0

#1004 - [Workaround] Puller polls s3 to ensure credentials

Pull Request - State: closed - Opened by perdasilva about 5 years ago

#1003 - [Feature] Show me all the currently running benchmarks...

Issue - State: open - Opened by gavinmbell about 5 years ago - 1 comment
Labels: enhancement, bai-bff, bai-client

#1002 - [bff][bugfix] tiny space bug stopping --sync-data flag from working

Pull Request - State: closed - Opened by gavinmbell about 5 years ago

#1001 - Add system logs to increase visibility

Issue - State: open - Opened by surajkota about 5 years ago - 2 comments
Labels: enhancement

#1000 - [Feature] Enhancement to list-jobs

Issue - State: open - Opened by surajkota about 5 years ago
Labels: enhancement, bai-bff, p2

#999 - TOML info for metric labels

Issue - State: open - Opened by gavinmbell about 5 years ago - 2 comments
Labels: RM:user-metrics

#998 - [SageMaker - exec] To pass more configuration information to SageMaker through TOML.

Issue - State: open - Opened by gavinmbell about 5 years ago - 1 comment
Labels: service:executor

#997 - No Limit for ES log query returns

Issue - State: open - Opened by gavinmbell about 5 years ago
Labels: enhancement, good first issue, bai-bff

#996 - [Improvement] Increase error state visibility to end users

Issue - State: open - Opened by perdasilva about 5 years ago - 3 comments
Labels: enhancement, good first issue

#995 - Patch GHSA-844w-j86r-4x2j

Pull Request - State: closed - Opened by Chancebair about 5 years ago - 1 comment

#994 - [Metrics] Improve UX with output.metrics regex's

Issue - State: open - Opened by jlcontreras about 5 years ago
Labels: enhancement, RM:user-metrics

#993 - Convert CloudWatch metrics to High resolution metrics

Pull Request - State: closed - Opened by jlcontreras about 5 years ago

#992 - [watcher] Add missing dependency

Pull Request - State: closed - Opened by jlcontreras about 5 years ago

#991 - Update metrics examples and information

Pull Request - State: closed - Opened by jlcontreras about 5 years ago - 1 comment

#990 - Update cmd event schema to include target-action-id field

Pull Request - State: closed - Opened by jlcontreras about 5 years ago - 1 comment

#989 - [Security] Updates external services to be annotated with the pl cidr ranges

Pull Request - State: closed - Opened by perdasilva about 5 years ago - 3 comments

#988 - Cast metric value to float before exporting to CloudWatch

Pull Request - State: closed - Opened by jlcontreras about 5 years ago

#984 - [Security] BFF/Grafana visible to the outside world

Issue - State: closed - Opened by perdasilva about 5 years ago - 2 comments
Labels: p0, security

#981 - User metrics doesn't show up in cloudwatch

Issue - State: closed - Opened by haohanchen-yagao over 5 years ago - 2 comments
Labels: p1

#980 - Single instance GPU job failed

Issue - State: open - Opened by akartsky over 5 years ago - 3 comments
Labels: p0

#976 - Apache Header All Of The Things!

Pull Request - State: closed - Opened by gavinmbell over 5 years ago - 6 comments

#910 - Deadletter Topic and mechanics

Issue - State: open - Opened by gavinmbell over 5 years ago
Labels: enhancement

#890 - [Customer Request] Trouble Running Script Mode Benchmark

Issue - State: open - Opened by Chancebair over 5 years ago - 4 comments
Labels: help wanted

#878 - [Executor] Refactor configuration

Issue - State: open - Opened by perdasilva over 5 years ago
Labels: enhancement, service:executor

#859 - Undeploy Not Idempotent

Issue - State: open - Opened by Chancebair over 5 years ago - 3 comments
Labels: bug, the build, p0

#856 - Blackbox test fail

Issue - State: closed - Opened by akhilmehra over 5 years ago - 1 comment

#846 - Blackbox Tests fail

Issue - State: closed - Opened by akhilmehra over 5 years ago - 1 comment
Labels: ci

#843 - [Watcher] Init:ImagePullBackOff - bad status

Issue - State: open - Opened by perdasilva over 5 years ago - 1 comment
Labels: bug, good first issue, service:watcher, p1

#816 - Fix the badges

Issue - State: open - Opened by stsukrov over 5 years ago - 1 comment
Labels: bug, ci

#792 - [CI] Blackbox Tests not utilizing the bff

Issue - State: open - Opened by Chancebair over 5 years ago - 1 comment
Labels: ci

#772 - Document Test setup

Issue - State: open - Opened by gavinmbell over 5 years ago - 1 comment
Labels: pa-handoff

#771 - Move code out of buildspec

Issue - State: open - Opened by gavinmbell over 5 years ago
Labels: pa-handoff, clean-up

#735 - Add "Manual Mode" to anubis-setup

Issue - State: open - Opened by Chancebair over 5 years ago - 1 comment
Labels: enhancement, ci, pa-handoff

#709 - [ci] Wait before get service bai-bff

Issue - State: open - Opened by Chancebair over 5 years ago - 1 comment

#702 - Fetcher deployment fails due to wrong kubectl version

Issue - State: open - Opened by marcoabreu over 5 years ago
Labels: service:datafetcher, ci, the build

#691 - Investigate if it's possible to make S3 Mock work as a public one.

Issue - State: open - Opened by stsukrov over 5 years ago
Labels: enhancement, clean-up

#593 - Parametrise metrics pusher backend in templates

Issue - State: open - Opened by stsukrov over 5 years ago - 1 comment
Labels: infrastructure, service:executor

#579 - Improve executor/src/transpiler/kubernetes_spec_logic.py

Issue - State: open - Opened by jlcontreras over 5 years ago
Labels: enhancement, service:executor, clean-up

#555 - Investigate consistensy issues in data-puller

Issue - State: closed - Opened by stsukrov over 5 years ago - 2 comments
Labels: bug, service:datafetcher

#538 - Add taints to `bai-worker` nodes

Issue - State: open - Opened by edisongustavo over 5 years ago - 2 comments
Labels: enhancement, infrastructure, p0

#453 - make deploy should not depend on make publish anymore

Issue - State: open - Opened by stsukrov over 5 years ago
Labels: enhancement, good first issue, ci

#401 - Executor randomly chooses AZ, but doesn't check whether it's suitable

Issue - State: open - Opened by jlcontreras over 5 years ago - 2 comments
Labels: bug, service:executor, p0, RM:improve-az-selection

#393 - FetcherStatus.FAILED message is empty

Issue - State: open - Opened by marcoabreu over 5 years ago - 1 comment
Labels: service:datafetcher, p2

#369 - Cancel jobs which require unavailable instance types

Issue - State: open - Opened by jlcontreras over 5 years ago - 1 comment
Labels: bug, service:executor

#344 - Adopt kustomize to manage different configurations (local/devo/prod)

Issue - State: open - Opened by stsukrov over 5 years ago - 1 comment
Labels: enhancement, baictl, clean-up

#326 - [BFF] Implement data validation (at ports and adaptors)

Issue - State: open - Opened by gavinmbell over 5 years ago - 4 comments
Labels: bai-bff

#320 - Upgrade to terraform 0.12

Issue - State: open - Opened by edisongustavo over 5 years ago - 3 comments
Labels: enhancement, infrastructure

#276 - Design bai scriptMode schanges in descriptor

Issue - State: open - Opened by stsukrov over 5 years ago
Labels: enhancement, descriptor, service:executor

#263 - Readlock from puller

Issue - State: open - Opened by stsukrov almost 6 years ago
Labels: enhancement, service:datafetcher

GitHub / awslabs/benchmark-ai issues and pull requests