HPE GreenLake for Machine Learning Development Environment SaaS ("Service") enables machine learning ("ML") engineers and researchers to focus on refining their models without the distraction of infrastructure complexities. It eliminates the need for boilerplate code associated with managing ML infrastructure, seamlessly integrates with various ML frameworks, and tools.

2.1. HPE GreenLake edge-to-cloud platform ("platform"): The Service runs on the platform (access to Service, access control and monitoring of Consumption).

2.2. Easy-and-Quick Machine Learning Infrastructure Setup: Eliminate operational hassle related to infrastructure setup, management, upgrades, and backups of Machine Learning Development Environment ("MLDE") software.

2.3. Efficiently Manage and Share Your AI Infrastructure: Efficiently manage and share your GPUs & accelerators with ML workflow-aware smart scheduling and resource management to improve productivity and collaboration for your ML development and operations teams.

2.4. Security: Execute workloads in your cloud environment, thus ensuring your models, code, and data remain secure within your environment. Use Role-Based Access Controls ("RBAC") to authorize development and MLOps teams to securely collaborate and share ML resources and artifacts.

2.5. Hybrid and Compatibility: Compatible with multiple customer-provided cloud vendors including AWS and GCP, with planned support for additional cloud vendors.

2.6. Seamless Scaling: Give teams the ability to train at any scale by managing machine provision, networking, data loading and fault tolerance, making distributed model training faster and easier.

3.1. Support: HPE Tech Care Service “Basic” service experience level.

4.1. Pre-Release Materials

HPE may make available to Customer certain software, features, functionality, improvements, and/or enhancements in advance of their general availability (Pre-Release Materials). Customer agrees the Pre-Release Materials: (i) are not to be used in a production environment; (ii) may or may not ever be made generally available by HPE as part of an update or otherwise; (iii) are not under warranty or support; (iv) are not at the level of compatibility, performance and/or scalability of the Service as the case may be; (v) may not operate correctly; and, (vi) may be subject to additional terms and conditions that are specific to such Pre-Release Materials. Customer agrees to notify HPE of any bugs, errors or problems with respect to Pre-Release Materials.

4.2. The Service will not provide any usage or usage estimation for Customer’s public cloud account.

4.3. The Service does not include troubleshooting with Customer for issues unrelated to the HPE Machine Learning Development Environment software.

5.1. Enable the installation of cloud Service components:

• 5.1.1. Customer will enable HPE to install the cloud software components of the Service.

5.2. User management:

• 5.2.1. Provide managed user identities including access permissions for organization and cluster level access.
• 5.2.2. Ensure necessary permissions and access levels are maintained.

5.3. Access and permissions:

• 5.3.1. When needed, provide necessary access and permissions to your AWS or GCP cloud tenant account for HPE’s team.

5.4. Data and model management:

• 5.4.1. Supply data and model artifacts for training and deployment, along with any specific requirements for access and integration.

5.5. Engagement and communication:

• 5.5.1. Engage in regular communication with our support and technical teams to address issues and align on project goals.

5.6. Compliance and Security:

• 5.6.1. Collaborate on security measures and compliance requirements relevant to your organization's policies and regulations to ensure adherence to security and compliance best practices.

5.7. Capacity Planning:

• 5.7.1. Provide insights into expected workloads and usage patterns to facilitate resource scaling and optimization.

6.1. Termination of a Subscription

The Service will notify the Customer of the approaching expiry of a Subscription. The Customer can purchase a follow-on Subscription before the current Subscription expires. The follow-on Subscription will start after the expiry of the current Subscription making the Service continuous.