Accelerate generative AI adoption: A managed service approach

[Cloud_Experts]

Hewlett Packard Enterprise has announced HPE Machine Learning Development Environment Software as a managed service on AWS and other cloud providers to help businesses transform to a generative-AI-enabled enterprise with a complete, adaptable cloud managed experience for AI/ML model training.

By Gary Craze, HPE Senior Product Marketing Manager, Artificial Intelligence and Machine Learning; Piyush Shukla, HPE Director of Artificial Intelligence and Machine Learning Product Marketing; and Ran Fu, HPE Group Product Manager

Artificial intelligence (AI) has rapidly evolved in recent years, with generative AI models at the forefront of innovation. These models, powered by deep learning algorithms, have the capability to generate human-like text, images, and even code. They hold tremendous potential in various domains, from content creation to automation and problem-solving.

However, adopting generative AI can be challenging for businesses due to the complexity of the technology, high computational requirements, and a shortage of skilled AI experts.

Introducing HPE Machine Learning Development Environment Software as a managed service on AWS and other cloud providers

As companies look toward transforming to a generative-AI-enabled enterprise, they face a variety of AI/ML development infrastructure challenges, such as complex integrations that require long lead times; the need for secure, flexible and futureproof technology to support growing performance and data demands; and the ability to keep support and maintenance streamlined. To help businesses meet these needs, HPE has announced HPE Machine Learning Development Environment Software as a managed service on AWS and other cloud providers.

The HPE Machine Learning Development Environment Software as a managed service helps customers transform to a generative-AI-enabled enterprise with a complete, adaptable cloud managed experience for AI/ML model training by helping them to: 

• Accelerate and securely implement generative AI initiatives as a force multiplier for their organization, in days not months, with a flexible managed service cloud option that supports any stage of a company’s AI/ML journey and adapts to their growth.  
• Reduce complexity and operational overheard by leveraging an AI/ML managed infrastructure option to accelerate the time to model development.  
• Future-proof their AI/ML infrastructure and help relieve management staffing and process burdens. 

Companies leverage the HPE Machine Learning Development Environment Software to help uncover hidden insights from their data by helping AI/ML engineers and data scientists collaborate, build more accurate ML models and train them faster.   

Now customers can benefit from accelerating their use of the HPE Machine Learning Development Environment through a managed service/cloud experience, enjoying the full feature set of HPE Machine Learning Development Environment as delivered on their existing AWS account.  

This flexible managed service features easy and quick infrastructure setup with support from HPE. It offers security and seamless scaling, allowing customers to begin using the HPE Machine Learning Development Environment Software often in a matter of days, as compared to longer, full on-premises deployments of AI infrastructure and software.  

Customers who wish to evaluate the functionality of the HPE Machine Learning Development Environment Software can take advantage of a free evaluation of the software in the managed service offering, and then transition to a fully licensed, paid, unlimited offering with a full SLA.   

The HPE Machine Learning Development Environment Software includes new features such as: 

• Managed Service - Bring Your Own Cloud (BYOC)

- Delivered on a customer’s existing cloud AWS. Easy and quick infrastructure setup with support from HPE.

• Extending the ML development enterprise platform:

- Enterprise-ready multi-tenancy

- Mix AI and HPC workloads on a shared cluster

- Accelerator heterogeneity

- Distributed training for Large Language Models

- Distributed batch inference

• Accelerating generative AI adoption

- Rapid prototyping and testing of generative AI models.

In addition to the core HPE Machine Learning Development Environment product, the managed service offering includes many additional services to further enable your teams to focus on what really matters, utilizing ML techniques and best practices to develop AI/ML models that scale. 

Included in the managed service: 

• Onboarding support and getting-started guidance with a partnered HPE engineer 
• Incident response and alerting to ensure platform performance and availability 
• Operational support and configuration including upgrades and backup management. 

For companies looking to transform to a generative-AI-enabled enterprise and wanting to mitigate the complexity of acquiring, deploying and managing complex AI infrastructure, the HPE Machine Learning Development Environment as a Managed Service can help deliver a secure, reliable and cost-efficient AI/ML technology and infrastructure solution, while accelerating their generative AI initiatives.  

Next step in the evolution towards a generative AI enterprise

Organizations face multiple challenges in leveraging the benefits of generative AI, including:

• Lack of generative AI model development expertise
• Developing a cost effective, iterative process of generative AI model development
• Understanding trade-offs between foundation model options for their application
• Moving models from PoC to production while navigating the cliffs of scalability and security.

To enable customers to achieve faster time to value with generative AI, HPE Machine Learning Development Environment is developing a generative AI studio feature, a guided, pre-defined generative AI model evaluation and development experience.

This new experience for your LLM journey will offer a choice of free, open-source foundation models, allowing you to go from model evaluation to model customization in hours, not days. On Day 1, you can simply chat with the open-source models like Llama 2, Falcon, and MBT. By self-hosting these models, you gain full control over your data, whether it is in a private or public cloud, thus ensuring data security and privacy.

Furthermore, the generative AI studio feature allows you to evaluate a model using simple prompts and gradually move to advanced techniques such as few-shot learning – all within a guided experience. The feature allows users to compare models based on quality, speed, and cost, enabling selection of the optimum solution. With prompt engineering your work can be saved and reproduced, tracked, and shared in a collaborative environment.

The ability to fine-tune models on your data is a simple process and the no-code experience allows you to start prototyping models immediately, without any LLM expertise. LLM tools like prompt engineering and fine-tuning are available for advanced users within the Notebook and experiment tracking dashboard.

Learn more about the HPE Machine Learning Development Environment Software as a service on AWS and other cloud providers.

For a free 90-day trial for HPE Machine Learning Development Environment: https://mldes.ext.hpe.com/trial

For more information on HPE Machine Learning Development Environment: www.hpe.com/hpe-machine-learning-development-software

Take the next step:

Meet us and see HPE Machine Learning Development Environment in action at AWS re:Invent Booth 630.

See HPE Machine Learning Development Environment at AWS re:Invent. Sign up for the session: “Fully managed AI Model Building: ML/DL Model Training at Scale on AWS”  

Tuesday, November 27, 2023 | 3:30 PM

MGM Grand Hotel, Chairman’s 370

See HPE Machine Learning Development Environment in action at HPE Discover Barcelona 2023:

DEMO603: Foundation models and RAG implementing practical GenAI

Session IS6353: Training & Tuning Generative AI for Enterprise Transformation

Thursday, November 30, 2023, 03:30 p.m. - 04:15 p.m.

Hall 8.0 Room 6

Gary Craze is a 35-year veteran of the technology industry. Gary has held marketing and product management roles with enterprise technology companies, helping them to understand the needs of customers and create compelling value propositions that meet their business needs. Currently, Gary is the Senior Product Marketing Manager for the HPE AI team, where he is responsible for evangelizing HPE’s leading AI software solutions.

Piyush Shukla is the Director of Artificial Intelligence and Machine Learning Product Marketing at HPE, responsible for driving critical and challenging AI and ML initiatives that support HPE's industry leadership in this space. During his 20-year career in high tech, he has been a frequent expert speaker at industry events, where he presented on topics such as analytics, hybrid cloud, container-based solutions, and VDI. Recognized as a visionary marketing leader, he has a solid reputation for delivering innovative products to HPE customers.

Ran Fu is Group Product Manager at Hewlett Packard Enterprise and is  leading the efforts to reinvent a new Gen AI platform to enable customers to quickly prototype, evaluate, customize, and deploy large language models. Ran is passionate about helping customers create more values with AI innovations. 

Cloud Services Experts
Hewlett Packard Enterprise

twitter.com/HPE_GreenLake
linkedin.com/showcase/hpe-greenlake/
hpe.com/us/en/greenlake