This part 3 from the European Mainframe Academy’s introduction to the IBM Z17 mainframe focuses on its integrated artificial intelligence capabilities. The discussion highlights the new Telum II processor, which significantly boosts on-chip AI acceleration for real-time inference directly within enterprise workloads. This integration allows for faster, more secure decision-making at the point of transaction, eliminating the need for offloading data. Furthermore, the Z17 provides tools for AI model management, observability, and automation, driving adoption in industries like finance and retail for use cases such as fraud detection and personalized offers.
The following video lasts 5 minutes. It’s worth watching!
A glossary of key terms mentioned in the video:
- AI (Artificial Intelligence): The theory and development of computer systems able to perform tasks that normally require human intelligence, such as visual perception, speech recognition, decision-making, and translation between languages.
- Telum II: The custom-designed processor in the IBM Z17 mainframe that includes an on-chip AI accelerator, representing a significant upgrade in AI compute power compared to its predecessor.
- Inference: The stage in machine learning where a trained model is used to make predictions or decisions based on new, unseen data.
- z/OS: The primary operating system for IBM mainframe computers.
- CICS (Customer Information Control System): A transaction processing system for IBM mainframes.
- IMS (Information Management System): A hierarchical database and transaction management system for IBM mainframes.
- DB2: A family of database management products from IBM, commonly used on mainframes.
- z/OS Connect: Enables the exposure of ZOS-based applications and data as RESTful APIs for hybrid cloud integration.
- REST Endpoint: A specific URL that represents a resource in a RESTful API, allowing for interaction and data exchange over a network.
- ONNX (Open Neural Network Exchange): An open standard format for representing machine learning models, facilitating interoperability between different AI frameworks.
- PyTorch: An open-source machine learning framework.
- TensorFlow: An open-source software library for numerical computation using data flow graphs, commonly used for machine learning.
- Watson Machine Learning for ZOS: IBM’s platform for building, deploying, and managing machine learning models on the ZOS operating system.
- ModelOps: A set of practices that aim to automate and standardize the deployment, monitoring, and governance of machine learning models.
- Observability: The ability to understand the internal state of a system based on its external outputs, often enhanced by AI to detect anomalies and predict issues.
- Automation: The use of technology to perform tasks with minimal human assistance.
- Root Cause Analysis: The process of identifying the underlying cause of a problem or failure.
- Latency: The delay between an input and the corresponding output or response.
- Data Egress Risk: The potential for sensitive data to leave a secure environment, posing a security threat.
- GPU (Graphics Processing Unit): A specialized electronic circuit designed to rapidly manipulate and alter memory to accelerate the creation of images in a frame buffer intended for output to a display device. Often used for accelerating machine learning model training.
- LLM (Large Language Model): A type of artificial intelligence model that is trained on a massive amount of text data to understand and generate human-like language.
- Granite Models: IBM’s family of large language models.
- Spyre: AI processing card that works in conjunction with Telum II to optimize hardware and software for leveraging LLMs on the z17.
- Parallel Sysplex: A clustering technology for IBM mainframes that enables multiple systems to operate as a single logical computing entity.
- GDPS (Geographically Dispersed Parallel Sysplex): An extension of Parallel Sysplex that provides disaster recovery capabilities by allowing the clustered systems to be located in geographically separate locations.
Recent Comments