기계 번역으로 제공되는 번역입니다. 제공된 번역과 원본 영어의 내용이 상충하는 경우에는 영어 버전이 우선합니다.

# TensorFlow
<a name="emr-tensorflow"></a>

TensorFlow는 머신 인텔리전스 및 딥 러닝 애플리케이션을 위한 오픈 소스 기호 수학 라이브러리입니다. 자세한 내용은 [TensorFlow 웹 사이트](https://www.tensorflow.org/)를 참조하세요. TensorFlow는 Amazon EMR 릴리스 버전 5.17.0 이상에서 사용할 수 있습니다.

다음 테이블에는 Amazon EMR이 TensorFlow를 통해 설치하는 구성 요소와 함께 Amazon EMR 7.x 시리즈의 최신 릴리스에 포함된 TensorFlow의 버전이 나열되어 있습니다.

이 릴리스에서 TensorFlow와 함께 설치된 구성 요소의 버전은 [릴리스 7.12.0 구성 요소 버전을 참조하세요](emr-7120-release.md).


**emr-7.12.0용 TensorFlow 버전 정보**  

| Amazon EMR 릴리스 레이블 | TensorFlow 버전 | TensorFlow와 함께 설치된 구성 요소 | 
| --- | --- | --- | 
| emr-7.12.0 | TensorFlow 2.19.0 | emrfs, emr-goodies, hadoop-client, hadoop-hdfs-datanode, hadoop-hdfs-library, hadoop-hdfs-namenode, hadoop-hdfs-zkfc, hadoop-httpfs-server, hadoop-kms-server, hadoop-yarn-nodemanager, hadoop-yarn-resourcemanager, hadoop-yarn-timeline-server, tensorflow | 

다음 테이블에는 Amazon EMR이 TensorFlow를 통해 설치하는 구성 요소와 함께 Amazon EMR 6.x 시리즈의 최신 릴리스에 포함된 TensorFlow의 버전이 나열되어 있습니다.

이 릴리스에서 TensorFlow와 함께 설치된 구성 요소의 버전은 [릴리스 6.15.0 구성 요소 버전](emr-6150-release.md)을 참조하세요.


**emr-6.15.0용 TensorFlow 버전 정보**  

| Amazon EMR 릴리스 레이블 | TensorFlow 버전 | TensorFlow와 함께 설치된 구성 요소 | 
| --- | --- | --- | 
| emr-6.15.0 | TensorFlow 2.11.0 | emrfs, emr-goodies, hadoop-client, hadoop-hdfs-datanode, hadoop-hdfs-library, hadoop-hdfs-namenode, hadoop-httpfs-server, hadoop-kms-server, hadoop-yarn-nodemanager, hadoop-yarn-resourcemanager, hadoop-yarn-timeline-server, tensorflow | 

다음 테이블에는 Amazon EMR이 TensorFlow를 통해 설치하는 구성 요소와 함께 Amazon EMR 5.x 시리즈의 최신 릴리스에 포함된 TensorFlow의 버전이 나열되어 있습니다.

이 릴리스에서 TensorFlow와 함께 설치된 구성 요소의 버전은 [릴리스 5.36.2 구성 요소 버전](emr-5362-release.md)을 참조하세요.


**emr-5.36.2용 TensorFlow 버전 정보**  

| Amazon EMR 릴리스 레이블 | TensorFlow 버전 | TensorFlow와 함께 설치된 구성 요소 | 
| --- | --- | --- | 
| emr-5.36.2 | TensorFlow 2.4.1 | emrfs, emr-goodies, hadoop-client, hadoop-hdfs-datanode, hadoop-hdfs-library, hadoop-hdfs-namenode, hadoop-httpfs-server, hadoop-kms-server, hadoop-yarn-nodemanager, hadoop-yarn-resourcemanager, hadoop-yarn-timeline-server, tensorflow | 

## Amazon EC2 인스턴스 유형별 TensorFlow 빌드
<a name="w2aac66c23"></a>

Amazon EMR은 클러스터에 대해 선택한 인스턴스 유형에 따라 다른 빌드의 TensorFlow 라이브러리를 사용합니다. 또한 Amazon EMR은 EMR-7.5.0 이상에서 aarch64 인스턴스 유형(Graviton 인스턴스)이 있는 클러스터에 TensorFlow를 지원합니다.

다음 표에는 EMR-7.10.0의 인스턴스 유형별 빌드가 나와 있습니다.


| EC2 인스턴스 유형 | TensorFlow 빌드 | 
| --- | --- | 
|  P2, P4D, P5, G4DN, G5, G6, GR6  |  CUDA 12.5, cuDNN 9.3.0.75가 적용된 Tensorflow 2.18.0  | 
|  P3, P3DN, G3, G3S  |  CUDA 12.5, cuDNN 9.3.0.75, NCCL 2.22.3이 적용된 Tensorflow 2.18.0 [Nvidia NCCL](https://developer.nvidia.com/nccl)은 버전 2.22.3의 P3, P3DN, G3 및 G3s 인스턴스에서만 사용할 수 있습니다. **최종 사용자 라이선스 계약(EULA)**: Amazon EMR에서 Nvidia 구성 요소를 사용하면 [제품 EULA](https://d7umqicpi7263.cloudfront.net/eula/product/d0199cf7-a04a-4204-be4d-dc3e2af678af/5b36dd71-7d6e-4d97-a8f7-013d3eccec70.txt)에 나와 있는 이용 약관에 동의하는 것입니다.  | 
|  Graviton 인스턴스  |  Tensorflow 2.18.0  | 
|  기타 모두  |  Tensorflow CPU 2.18.0  | 

## 보안
<a name="w2aac66c25"></a>

[Using TensorFlow securely](https://github.com/tensorflow/tensorflow/blob/master/SECURITY.md)의 지침을 따르고 프라이빗 서브넷에서 클러스터를 시작하여 신뢰할 수 있는 출처로 액세스를 제한하는 것이 좋습니다. 자세한 내용은 *Amazon EMR 관리 안내서*에서 [Amazon VPC 옵션](https://docs.aws.amazon.com/emr/latest/ManagementGuide/emr-clusters-in-a-vpc.html#emr-vpc-private-subnet)을 참조하세요.

## TensorBoard 사용
<a name="emr-tensorflow-tensorboard"></a>

TensorBoard는 TensorFlow 프로그램용 시각화 도구 제품군입니다. 자세한 내용은 Tensorflow 웹 사이트에서 [TensorBoard: Visualized learning](https://www.tensorflow.org/get_started/summaries_and_tensorboard)을 참조하세요.

TensorBoard를 Amazon EMR과 함께 사용하려면 클러스터 프라이머리 노드에서 TensorBoard를 시작해야 합니다.

**Amazon EMR에서 Tensorboard를 Tensorflow와 함께 사용하는 방법**

1. SSH를 사용하여 클러스터의 마스터 노드에 연결합니다. 자세한 내용은 *Amazon EMR 관리 안내서*에서 [SSH를 사용하여 프라이머리 노드에 연결](https://docs.aws.amazon.com/emr/latest/ManagementGuide/emr-connect-master-node-ssh.html)을 참조하세요.

1. 다음 명령을 입력하여 마스터 노드에서 Tensorboard를 시작합니다. 요약 쓰기를 사용하여 요약 데이터를 생성하고 저장한 마스터 노드의 디렉터리로 `/my/log/directory`를 바꿉니다.

------
#### [ Amazon EMR 5.19.0 and later ]

   ```
   python3 -m tensorboard.main --logdir=/home/hadoop/tensor --bind_all
   ```

------
#### [ Amazon EMR 5.18.1 and earlier ]

   ```
   python3 -m tensorboard.main --logdir=/my/log/dir
   ```

------

   기본적으로 마스터 노드는 포트 6006과 마스터 퍼블릭 DNS 이름을 사용하여 TensorBoard를 호스팅합니다. TensorBoard를 시작한 후 다음 예제와 같이 TensorBoard에 연결하는 데 사용할 수 있는 URL이 명령줄 출력에 표시됩니다.

   ```
   TensorBoard 2.18.0 at http://master-public-dns-name:6006 (Press CTRL+C to quit)
   ```

1. 신뢰할 수 있는 클라이언트의 마스터 노드에서 웹 인터페이스에 액세스하도록 설정합니다. 자세한 내용은 *Amazon EMR 관리 안내서*에서 [Amazon EMR 클러스터에 호스팅된 웹 인터페이스 보기](https://docs.aws.amazon.com/emr/latest/ManagementGuide/emr-web-interfaces.html)를 참조하세요.

1. `http://master-public-dns-name:6006`에서 TensorBoard를 엽니다.

# TensorFlow 릴리스 기록
<a name="TensorFlow-release-history"></a>

다음 테이블에는 애플리케이션과 함께 설치된 구성 요소 외에도 Amazon EMR의 최신 릴리스에 포함된 TensorFlow의 버전이 나열되어 있습니다. 각 릴리스의 구성 요소 버전은 [Amazon EMR 7.x 릴리스 버전](emr-release-7x.md), [Amazon EMR 6.x 릴리스 버전](emr-release-6x.md) 또는 [Amazon EMR 5.x 릴리스 버전](emr-release-5x.md)의 릴리스의 구성 요소 버전 섹션을 참조하세요.


**TensorFlow 버전 정보**  

| Amazon EMR 릴리스 레이블 | TensorFlow 버전 | TensorFlow와 함께 설치된 구성 요소 | 
| --- | --- | --- | 
| emr-7.12.0 | 2.19.0 | emrfs, emr-goodies, hadoop-client, hadoop-hdfs-datanode, hadoop-hdfs-library, hadoop-hdfs-namenode, hadoop-hdfs-zkfc, hadoop-httpfs-server, hadoop-kms-server, hadoop-yarn-nodemanager, hadoop-yarn-resourcemanager, hadoop-yarn-timeline-server, tensorflow | 
| emr-7.11.0 | 2.19.0 | emrfs, emr-goodies, hadoop-client, hadoop-hdfs-datanode, hadoop-hdfs-library, hadoop-hdfs-namenode, hadoop-hdfs-zkfc, hadoop-httpfs-server, hadoop-kms-server, hadoop-yarn-nodemanager, hadoop-yarn-resourcemanager, hadoop-yarn-timeline-server, tensorflow | 
| emr-7.10.0 | 2.18.0 | emrfs, emr-goodies, hadoop-client, hadoop-hdfs-datanode, hadoop-hdfs-library, hadoop-hdfs-namenode, hadoop-httpfs-server, hadoop-kms-server, hadoop-yarn-nodemanager, hadoop-yarn-resourcemanager, hadoop-yarn-timeline-server, tensorflow | 
| emr-7.9.0 | 2.16.1 | emrfs, emr-goodies, hadoop-client, hadoop-hdfs-datanode, hadoop-hdfs-library, hadoop-hdfs-namenode, hadoop-httpfs-server, hadoop-kms-server, hadoop-yarn-nodemanager, hadoop-yarn-resourcemanager, hadoop-yarn-timeline-server, tensorflow | 
| emr-7.8.0 | 2.16.1 | emrfs, emr-goodies, hadoop-client, hadoop-hdfs-datanode, hadoop-hdfs-library, hadoop-hdfs-namenode, hadoop-httpfs-server, hadoop-kms-server, hadoop-yarn-nodemanager, hadoop-yarn-resourcemanager, hadoop-yarn-timeline-server, tensorflow | 
| emr-7.7.0 | 2.16.1 | emrfs, emr-goodies, hadoop-client, hadoop-hdfs-datanode, hadoop-hdfs-library, hadoop-hdfs-namenode, hadoop-httpfs-server, hadoop-kms-server, hadoop-yarn-nodemanager, hadoop-yarn-resourcemanager, hadoop-yarn-timeline-server, tensorflow | 
| emr-7.6.0 | 2.16.1 | emrfs, emr-goodies, hadoop-client, hadoop-hdfs-datanode, hadoop-hdfs-library, hadoop-hdfs-namenode, hadoop-httpfs-server, hadoop-kms-server, hadoop-yarn-nodemanager, hadoop-yarn-resourcemanager, hadoop-yarn-timeline-server, tensorflow | 
| emr-7.5.0 | 2.16.1 | emrfs, emr-goodies, hadoop-client, hadoop-hdfs-datanode, hadoop-hdfs-library, hadoop-hdfs-namenode, hadoop-httpfs-server, hadoop-kms-server, hadoop-yarn-nodemanager, hadoop-yarn-resourcemanager, hadoop-yarn-timeline-server, tensorflow | 
| emr-7.4.0 | 2.16.1 | emrfs, emr-goodies, hadoop-client, hadoop-hdfs-datanode, hadoop-hdfs-library, hadoop-hdfs-namenode, hadoop-httpfs-server, hadoop-kms-server, hadoop-yarn-nodemanager, hadoop-yarn-resourcemanager, hadoop-yarn-timeline-server, tensorflow | 
| emr-7.3.0 | 2.16.1 | emrfs, emr-goodies, hadoop-client, hadoop-hdfs-datanode, hadoop-hdfs-library, hadoop-hdfs-namenode, hadoop-httpfs-server, hadoop-kms-server, hadoop-yarn-nodemanager, hadoop-yarn-resourcemanager, hadoop-yarn-timeline-server, tensorflow | 
| emr-7.2.0 | 2.11.0 | emrfs, emr-goodies, hadoop-client, hadoop-hdfs-datanode, hadoop-hdfs-library, hadoop-hdfs-namenode, hadoop-httpfs-server, hadoop-kms-server, hadoop-yarn-nodemanager, hadoop-yarn-resourcemanager, hadoop-yarn-timeline-server, tensorflow | 
| emr-5.36.2 | 2.4.1 | emrfs, emr-goodies, hadoop-client, hadoop-hdfs-datanode, hadoop-hdfs-library, hadoop-hdfs-namenode, hadoop-httpfs-server, hadoop-kms-server, hadoop-yarn-nodemanager, hadoop-yarn-resourcemanager, hadoop-yarn-timeline-server, tensorflow | 
| emr-7.1.0 | 2.11.0 | emrfs, emr-goodies, hadoop-client, hadoop-hdfs-datanode, hadoop-hdfs-library, hadoop-hdfs-namenode, hadoop-httpfs-server, hadoop-kms-server, hadoop-yarn-nodemanager, hadoop-yarn-resourcemanager, hadoop-yarn-timeline-server, tensorflow | 
| emr-7.0.0 | 2.11.0 | emrfs, emr-goodies, hadoop-client, hadoop-hdfs-datanode, hadoop-hdfs-library, hadoop-hdfs-namenode, hadoop-httpfs-server, hadoop-kms-server, hadoop-yarn-nodemanager, hadoop-yarn-resourcemanager, hadoop-yarn-timeline-server, tensorflow | 
| emr-6.15.0 | 2.11.0 | emrfs, emr-goodies, hadoop-client, hadoop-hdfs-datanode, hadoop-hdfs-library, hadoop-hdfs-namenode, hadoop-httpfs-server, hadoop-kms-server, hadoop-yarn-nodemanager, hadoop-yarn-resourcemanager, hadoop-yarn-timeline-server, tensorflow | 
| emr-6.14.0 | 2.11.0 | emrfs, emr-goodies, hadoop-client, hadoop-hdfs-datanode, hadoop-hdfs-library, hadoop-hdfs-namenode, hadoop-httpfs-server, hadoop-kms-server, hadoop-yarn-nodemanager, hadoop-yarn-resourcemanager, hadoop-yarn-timeline-server, tensorflow | 
| emr-6.13.0 | 2.11.0 | emrfs, emr-goodies, hadoop-client, hadoop-hdfs-datanode, hadoop-hdfs-library, hadoop-hdfs-namenode, hadoop-httpfs-server, hadoop-kms-server, hadoop-yarn-nodemanager, hadoop-yarn-resourcemanager, hadoop-yarn-timeline-server, tensorflow | 
| emr-6.12.0 | 2.11.0 | emrfs, emr-goodies, hadoop-client, hadoop-hdfs-datanode, hadoop-hdfs-library, hadoop-hdfs-namenode, hadoop-httpfs-server, hadoop-kms-server, hadoop-yarn-nodemanager, hadoop-yarn-resourcemanager, hadoop-yarn-timeline-server, tensorflow | 
| emr-6.11.1 | 2.11.0 | emrfs, emr-goodies, hadoop-client, hadoop-hdfs-datanode, hadoop-hdfs-library, hadoop-hdfs-namenode, hadoop-httpfs-server, hadoop-kms-server, hadoop-yarn-nodemanager, hadoop-yarn-resourcemanager, hadoop-yarn-timeline-server, tensorflow | 
| emr-6.11.0 | 2.11.0 | emrfs, emr-goodies, hadoop-client, hadoop-hdfs-datanode, hadoop-hdfs-library, hadoop-hdfs-namenode, hadoop-httpfs-server, hadoop-kms-server, hadoop-yarn-nodemanager, hadoop-yarn-resourcemanager, hadoop-yarn-timeline-server, tensorflow | 
| emr-6.10.1 | 2.11.0 | emrfs, emr-goodies, hadoop-client, hadoop-hdfs-datanode, hadoop-hdfs-library, hadoop-hdfs-namenode, hadoop-httpfs-server, hadoop-kms-server, hadoop-yarn-nodemanager, hadoop-yarn-resourcemanager, hadoop-yarn-timeline-server, tensorflow | 
| emr-6.10.0 | 2.11.0 | emrfs, emr-goodies, hadoop-client, hadoop-hdfs-datanode, hadoop-hdfs-library, hadoop-hdfs-namenode, hadoop-httpfs-server, hadoop-kms-server, hadoop-yarn-nodemanager, hadoop-yarn-resourcemanager, hadoop-yarn-timeline-server, tensorflow | 
| emr-6.9.1 | 2.10.0 | emrfs, emr-goodies, hadoop-client, hadoop-hdfs-datanode, hadoop-hdfs-library, hadoop-hdfs-namenode, hadoop-httpfs-server, hadoop-kms-server, hadoop-yarn-nodemanager, hadoop-yarn-resourcemanager, hadoop-yarn-timeline-server, tensorflow | 
| emr-6.9.0 | 2.10.0 | emrfs, emr-goodies, hadoop-client, hadoop-hdfs-datanode, hadoop-hdfs-library, hadoop-hdfs-namenode, hadoop-httpfs-server, hadoop-kms-server, hadoop-yarn-nodemanager, hadoop-yarn-resourcemanager, hadoop-yarn-timeline-server, tensorflow | 
| emr-6.8.1 | 2.9.1 | emrfs, emr-goodies, hadoop-client, hadoop-hdfs-datanode, hadoop-hdfs-library, hadoop-hdfs-namenode, hadoop-httpfs-server, hadoop-kms-server, hadoop-yarn-nodemanager, hadoop-yarn-resourcemanager, hadoop-yarn-timeline-server, tensorflow | 
| emr-6.8.0 | 2.9.1 | emrfs, emr-goodies, hadoop-client, hadoop-hdfs-datanode, hadoop-hdfs-library, hadoop-hdfs-namenode, hadoop-httpfs-server, hadoop-kms-server, hadoop-yarn-nodemanager, hadoop-yarn-resourcemanager, hadoop-yarn-timeline-server, tensorflow | 
| emr-6.7.0 | 2.4.1 | emrfs, emr-goodies, hadoop-client, hadoop-hdfs-datanode, hadoop-hdfs-library, hadoop-hdfs-namenode, hadoop-httpfs-server, hadoop-kms-server, hadoop-yarn-nodemanager, hadoop-yarn-resourcemanager, hadoop-yarn-timeline-server, tensorflow | 
| emr-5.36.1 | 2.4.1 | emrfs, emr-goodies, hadoop-client, hadoop-hdfs-datanode, hadoop-hdfs-library, hadoop-hdfs-namenode, hadoop-httpfs-server, hadoop-kms-server, hadoop-yarn-nodemanager, hadoop-yarn-resourcemanager, hadoop-yarn-timeline-server, tensorflow | 
| emr-5.36.0 | 2.4.1 | emrfs, emr-goodies, hadoop-client, hadoop-hdfs-datanode, hadoop-hdfs-library, hadoop-hdfs-namenode, hadoop-httpfs-server, hadoop-kms-server, hadoop-yarn-nodemanager, hadoop-yarn-resourcemanager, hadoop-yarn-timeline-server, tensorflow | 
| emr-6.6.0 | 2.4.1 | emrfs, emr-goodies, hadoop-client, hadoop-hdfs-datanode, hadoop-hdfs-library, hadoop-hdfs-namenode, hadoop-httpfs-server, hadoop-kms-server, hadoop-yarn-nodemanager, hadoop-yarn-resourcemanager, hadoop-yarn-timeline-server, tensorflow | 
| emr-5.35.0 | 2.4.1 | emrfs, emr-goodies, hadoop-client, hadoop-hdfs-datanode, hadoop-hdfs-library, hadoop-hdfs-namenode, hadoop-httpfs-server, hadoop-kms-server, hadoop-yarn-nodemanager, hadoop-yarn-resourcemanager, hadoop-yarn-timeline-server, tensorflow | 
| emr-6.5.0 | 2.4.1 | emrfs, emr-goodies, hadoop-client, hadoop-hdfs-datanode, hadoop-hdfs-library, hadoop-hdfs-namenode, hadoop-httpfs-server, hadoop-kms-server, hadoop-yarn-nodemanager, hadoop-yarn-resourcemanager, hadoop-yarn-timeline-server, tensorflow | 
| emr-6.4.0 | 2.4.1 | emrfs, emr-goodies, hadoop-client, hadoop-hdfs-datanode, hadoop-hdfs-library, hadoop-hdfs-namenode, hadoop-httpfs-server, hadoop-kms-server, hadoop-yarn-nodemanager, hadoop-yarn-resourcemanager, hadoop-yarn-timeline-server, tensorflow | 
| emr-6.3.1 | 2.4.1 | emrfs, emr-goodies, hadoop-client, hadoop-hdfs-datanode, hadoop-hdfs-library, hadoop-hdfs-namenode, hadoop-httpfs-server, hadoop-kms-server, hadoop-yarn-nodemanager, hadoop-yarn-resourcemanager, hadoop-yarn-timeline-server, tensorflow | 
| emr-6.3.0 | 2.4.1 | emrfs, emr-goodies, hadoop-client, hadoop-hdfs-datanode, hadoop-hdfs-library, hadoop-hdfs-namenode, hadoop-httpfs-server, hadoop-kms-server, hadoop-yarn-nodemanager, hadoop-yarn-resourcemanager, hadoop-yarn-timeline-server, tensorflow | 
| emr-6.2.1 | 2.3.1 | emrfs, emr-goodies, hadoop-client, hadoop-hdfs-datanode, hadoop-hdfs-library, hadoop-hdfs-namenode, hadoop-httpfs-server, hadoop-kms-server, hadoop-yarn-nodemanager, hadoop-yarn-resourcemanager, hadoop-yarn-timeline-server, tensorflow | 
| emr-6.2.0 | 2.3.1 | emrfs, emr-goodies, hadoop-client, hadoop-hdfs-datanode, hadoop-hdfs-library, hadoop-hdfs-namenode, hadoop-httpfs-server, hadoop-kms-server, hadoop-yarn-nodemanager, hadoop-yarn-resourcemanager, hadoop-yarn-timeline-server, tensorflow | 
| emr-6.1.1 | 2.1.0 | emrfs, emr-goodies, hadoop-client, hadoop-hdfs-datanode, hadoop-hdfs-library, hadoop-hdfs-namenode, hadoop-httpfs-server, hadoop-kms-server, hadoop-yarn-nodemanager, hadoop-yarn-resourcemanager, hadoop-yarn-timeline-server, tensorflow | 
| emr-6.1.0 | 2.1.0 | emrfs, emr-goodies, hadoop-client, hadoop-hdfs-datanode, hadoop-hdfs-library, hadoop-hdfs-namenode, hadoop-httpfs-server, hadoop-kms-server, hadoop-yarn-nodemanager, hadoop-yarn-resourcemanager, hadoop-yarn-timeline-server, tensorflow | 
| emr-6.0.1 | 1.14.0 | emrfs, emr-goodies, hadoop-client, hadoop-hdfs-datanode, hadoop-hdfs-library, hadoop-hdfs-namenode, hadoop-httpfs-server, hadoop-kms-server, hadoop-yarn-nodemanager, hadoop-yarn-resourcemanager, hadoop-yarn-timeline-server, tensorflow | 
| emr-6.0.0 | 1.14.0 | emrfs, emr-goodies, hadoop-client, hadoop-hdfs-datanode, hadoop-hdfs-library, hadoop-hdfs-namenode, hadoop-httpfs-server, hadoop-kms-server, hadoop-yarn-nodemanager, hadoop-yarn-resourcemanager, hadoop-yarn-timeline-server, tensorflow | 
| emr-5.34.0 | 2.4.1 | emrfs, emr-goodies, hadoop-client, hadoop-hdfs-datanode, hadoop-hdfs-library, hadoop-hdfs-namenode, hadoop-httpfs-server, hadoop-kms-server, hadoop-yarn-nodemanager, hadoop-yarn-resourcemanager, hadoop-yarn-timeline-server, tensorflow | 
| emr-5.33.1 | 2.4.1 | emrfs, emr-goodies, hadoop-client, hadoop-hdfs-datanode, hadoop-hdfs-library, hadoop-hdfs-namenode, hadoop-httpfs-server, hadoop-kms-server, hadoop-yarn-nodemanager, hadoop-yarn-resourcemanager, hadoop-yarn-timeline-server, tensorflow | 
| emr-5.33.0 | 2.4.1 | emrfs, emr-goodies, hadoop-client, hadoop-hdfs-datanode, hadoop-hdfs-library, hadoop-hdfs-namenode, hadoop-httpfs-server, hadoop-kms-server, hadoop-yarn-nodemanager, hadoop-yarn-resourcemanager, hadoop-yarn-timeline-server, tensorflow | 
| emr-5.32.1 | 2.3.1 | emrfs, emr-goodies, hadoop-client, hadoop-hdfs-datanode, hadoop-hdfs-library, hadoop-hdfs-namenode, hadoop-httpfs-server, hadoop-kms-server, hadoop-yarn-nodemanager, hadoop-yarn-resourcemanager, hadoop-yarn-timeline-server, tensorflow | 
| emr-5.32.0 | 2.3.1 | emrfs, emr-goodies, hadoop-client, hadoop-hdfs-datanode, hadoop-hdfs-library, hadoop-hdfs-namenode, hadoop-httpfs-server, hadoop-kms-server, hadoop-yarn-nodemanager, hadoop-yarn-resourcemanager, hadoop-yarn-timeline-server, tensorflow | 
| emr-5.31.1 | 2.1.0 | emrfs, emr-goodies, hadoop-client, hadoop-hdfs-datanode, hadoop-hdfs-library, hadoop-hdfs-namenode, hadoop-httpfs-server, hadoop-kms-server, hadoop-yarn-nodemanager, hadoop-yarn-resourcemanager, hadoop-yarn-timeline-server, tensorflow | 
| emr-5.31.0 | 2.1.0 | emrfs, emr-goodies, hadoop-client, hadoop-hdfs-datanode, hadoop-hdfs-library, hadoop-hdfs-namenode, hadoop-httpfs-server, hadoop-kms-server, hadoop-yarn-nodemanager, hadoop-yarn-resourcemanager, hadoop-yarn-timeline-server, tensorflow | 
| emr-5.30.2 | 1.14.0 | emrfs, emr-goodies, hadoop-client, hadoop-hdfs-datanode, hadoop-hdfs-library, hadoop-hdfs-namenode, hadoop-httpfs-server, hadoop-kms-server, hadoop-yarn-nodemanager, hadoop-yarn-resourcemanager, hadoop-yarn-timeline-server, tensorflow | 
| emr-5.30.1 | 1.14.0 | emrfs, emr-goodies, hadoop-client, hadoop-hdfs-datanode, hadoop-hdfs-library, hadoop-hdfs-namenode, hadoop-httpfs-server, hadoop-kms-server, hadoop-yarn-nodemanager, hadoop-yarn-resourcemanager, hadoop-yarn-timeline-server, tensorflow | 
| emr-5.30.0 | 1.14.0 | emrfs, emr-goodies, hadoop-client, hadoop-hdfs-datanode, hadoop-hdfs-library, hadoop-hdfs-namenode, hadoop-httpfs-server, hadoop-kms-server, hadoop-yarn-nodemanager, hadoop-yarn-resourcemanager, hadoop-yarn-timeline-server, tensorflow | 
| emr-5.29.0 | 1.14.0 | emrfs, emr-goodies, hadoop-client, hadoop-hdfs-datanode, hadoop-hdfs-library, hadoop-hdfs-namenode, hadoop-httpfs-server, hadoop-kms-server, hadoop-yarn-nodemanager, hadoop-yarn-resourcemanager, hadoop-yarn-timeline-server, tensorflow | 
| emr-5.28.1 | 1.14.0 | emrfs, emr-goodies, hadoop-client, hadoop-hdfs-datanode, hadoop-hdfs-library, hadoop-hdfs-namenode, hadoop-httpfs-server, hadoop-kms-server, hadoop-yarn-nodemanager, hadoop-yarn-resourcemanager, hadoop-yarn-timeline-server, tensorflow | 
| emr-5.28.0 | 1.14.0 | emrfs, emr-goodies, hadoop-client, hadoop-hdfs-datanode, hadoop-hdfs-library, hadoop-hdfs-namenode, hadoop-httpfs-server, hadoop-kms-server, hadoop-yarn-nodemanager, hadoop-yarn-resourcemanager, hadoop-yarn-timeline-server, tensorflow | 
| emr-5.27.1 | 1.14.0 | emrfs, emr-goodies, hadoop-client, hadoop-hdfs-datanode, hadoop-hdfs-library, hadoop-hdfs-namenode, hadoop-httpfs-server, hadoop-kms-server, hadoop-yarn-nodemanager, hadoop-yarn-resourcemanager, hadoop-yarn-timeline-server, tensorflow | 
| emr-5.27.0 | 1.14.0 | emrfs, emr-goodies, hadoop-client, hadoop-hdfs-datanode, hadoop-hdfs-library, hadoop-hdfs-namenode, hadoop-httpfs-server, hadoop-kms-server, hadoop-yarn-nodemanager, hadoop-yarn-resourcemanager, hadoop-yarn-timeline-server, tensorflow | 
| emr-5.26.0 | 1.13.1 | emrfs, emr-goodies, hadoop-client, hadoop-hdfs-datanode, hadoop-hdfs-library, hadoop-hdfs-namenode, hadoop-httpfs-server, hadoop-kms-server, hadoop-yarn-nodemanager, hadoop-yarn-resourcemanager, hadoop-yarn-timeline-server, tensorflow | 
| emr-5.25.0 | 1.13.1 | emrfs, emr-goodies, hadoop-client, hadoop-hdfs-datanode, hadoop-hdfs-library, hadoop-hdfs-namenode, hadoop-httpfs-server, hadoop-kms-server, hadoop-yarn-nodemanager, hadoop-yarn-resourcemanager, hadoop-yarn-timeline-server, tensorflow | 
| emr-5.24.1 | 1.12.0 | emrfs, emr-goodies, hadoop-client, hadoop-hdfs-datanode, hadoop-hdfs-library, hadoop-hdfs-namenode, hadoop-httpfs-server, hadoop-kms-server, hadoop-yarn-nodemanager, hadoop-yarn-resourcemanager, hadoop-yarn-timeline-server, tensorflow | 
| emr-5.24.0 | 1.12.0 | emrfs, emr-goodies, hadoop-client, hadoop-hdfs-datanode, hadoop-hdfs-library, hadoop-hdfs-namenode, hadoop-httpfs-server, hadoop-kms-server, hadoop-yarn-nodemanager, hadoop-yarn-resourcemanager, hadoop-yarn-timeline-server, tensorflow | 
| emr-5.23.1 | 1.12.0 | emrfs, emr-goodies, hadoop-client, hadoop-hdfs-datanode, hadoop-hdfs-library, hadoop-hdfs-namenode, hadoop-httpfs-server, hadoop-kms-server, hadoop-yarn-nodemanager, hadoop-yarn-resourcemanager, hadoop-yarn-timeline-server, tensorflow | 
| emr-5.23.0 | 1.12.0 | emrfs, emr-goodies, hadoop-client, hadoop-hdfs-datanode, hadoop-hdfs-library, hadoop-hdfs-namenode, hadoop-httpfs-server, hadoop-kms-server, hadoop-yarn-nodemanager, hadoop-yarn-resourcemanager, hadoop-yarn-timeline-server, tensorflow | 
| emr-5.22.0 | 1.12.0 | emrfs, emr-goodies, hadoop-client, hadoop-hdfs-datanode, hadoop-hdfs-library, hadoop-hdfs-namenode, hadoop-httpfs-server, hadoop-kms-server, hadoop-yarn-nodemanager, hadoop-yarn-resourcemanager, hadoop-yarn-timeline-server, tensorflow | 
| emr-5.21.2 | 1.12.0 | emrfs, emr-goodies, hadoop-client, hadoop-hdfs-datanode, hadoop-hdfs-library, hadoop-hdfs-namenode, hadoop-httpfs-server, hadoop-kms-server, hadoop-yarn-nodemanager, hadoop-yarn-resourcemanager, hadoop-yarn-timeline-server, tensorflow | 
| emr-5.21.1 | 1.12.0 | emrfs, emr-goodies, hadoop-client, hadoop-hdfs-datanode, hadoop-hdfs-library, hadoop-hdfs-namenode, hadoop-httpfs-server, hadoop-kms-server, hadoop-yarn-nodemanager, hadoop-yarn-resourcemanager, hadoop-yarn-timeline-server, tensorflow | 
| emr-5.21.0 | 1.12.0 | emrfs, emr-goodies, hadoop-client, hadoop-hdfs-datanode, hadoop-hdfs-library, hadoop-hdfs-namenode, hadoop-httpfs-server, hadoop-kms-server, hadoop-yarn-nodemanager, hadoop-yarn-resourcemanager, hadoop-yarn-timeline-server, tensorflow | 
| emr-5.20.1 | 1.12.0 | emrfs, emr-goodies, hadoop-client, hadoop-hdfs-datanode, hadoop-hdfs-library, hadoop-hdfs-namenode, hadoop-httpfs-server, hadoop-kms-server, hadoop-yarn-nodemanager, hadoop-yarn-resourcemanager, hadoop-yarn-timeline-server, tensorflow | 
| emr-5.20.0 | 1.12.0 | emrfs, emr-goodies, hadoop-client, hadoop-hdfs-datanode, hadoop-hdfs-library, hadoop-hdfs-namenode, hadoop-httpfs-server, hadoop-kms-server, hadoop-yarn-nodemanager, hadoop-yarn-resourcemanager, hadoop-yarn-timeline-server, tensorflow | 
| emr-5.19.1 | 1.11.0 | emrfs, emr-goodies, hadoop-client, hadoop-hdfs-datanode, hadoop-hdfs-library, hadoop-hdfs-namenode, hadoop-httpfs-server, hadoop-kms-server, hadoop-yarn-nodemanager, hadoop-yarn-resourcemanager, hadoop-yarn-timeline-server, tensorflow | 
| emr-5.19.0 | 1.11.0 | emrfs, emr-goodies, hadoop-client, hadoop-hdfs-datanode, hadoop-hdfs-library, hadoop-hdfs-namenode, hadoop-httpfs-server, hadoop-kms-server, hadoop-yarn-nodemanager, hadoop-yarn-resourcemanager, hadoop-yarn-timeline-server, tensorflow | 
| emr-5.18.1 | 1.9.0 | emrfs, emr-goodies, hadoop-client, hadoop-hdfs-datanode, hadoop-hdfs-library, hadoop-hdfs-namenode, hadoop-httpfs-server, hadoop-kms-server, hadoop-yarn-nodemanager, hadoop-yarn-resourcemanager, hadoop-yarn-timeline-server, tensorflow | 
| emr-5.18.0 | 1.9.0 | emrfs, emr-goodies, hadoop-client, hadoop-hdfs-datanode, hadoop-hdfs-library, hadoop-hdfs-namenode, hadoop-httpfs-server, hadoop-kms-server, hadoop-yarn-nodemanager, hadoop-yarn-resourcemanager, hadoop-yarn-timeline-server, tensorflow | 
| emr-5.17.2 | 1.9.0 | emrfs, emr-goodies, hadoop-client, hadoop-hdfs-datanode, hadoop-hdfs-library, hadoop-hdfs-namenode, hadoop-httpfs-server, hadoop-kms-server, hadoop-yarn-nodemanager, hadoop-yarn-resourcemanager, hadoop-yarn-timeline-server, tensorflow | 
| emr-5.17.1 | 1.9.0 | emrfs, emr-goodies, hadoop-client, hadoop-hdfs-datanode, hadoop-hdfs-library, hadoop-hdfs-namenode, hadoop-httpfs-server, hadoop-kms-server, hadoop-yarn-nodemanager, hadoop-yarn-resourcemanager, hadoop-yarn-timeline-server, tensorflow | 
| emr-5.17.0 | 1.9.0 | emrfs, emr-goodies, hadoop-client, hadoop-hdfs-datanode, hadoop-hdfs-library, hadoop-hdfs-namenode, hadoop-httpfs-server, hadoop-kms-server, hadoop-yarn-nodemanager, hadoop-yarn-resourcemanager, hadoop-yarn-timeline-server, tensorflow |