Supporting IT infrastructure

Supporting IT infrastructure: a broader computational environment, consisting of servers, networking, database services, cloud services, applications. The IT infrastructure enables auxiliary operations, needed for operation of the target machine learning model. In the portrayed case, it enables the storage and provisioning of input data embeddings for inference operations and machine learning operations. As LLM’s are resource demanding models, the infrastructure may be structured to enable high-performance networking and scalable storage solutions. This will enable swift communication between computing and storage resources and will accommodate large storage and operation with large datasets. There can be multiple of IT infrastructure components “*”, which enable operation of one machine learning model.

Defined asset’s methods:

1. enableMachineLearningModelOperations() – the IT infrastructure accommodates the communication between computing resources to enable execution of processes, including inference operations of the machine learning model.
2. storeMLEmbeddings() – storage of input data embeddings business assets for later additional processing.

Involved business assets in association between the system asset and machine learning model:

• Input data embeddings: numerical vector representation of the input textual data, created either during the training or inference processes. This representation allows machine learning models, including LLM’s, to conduct defined calculations on the data, determine semantic relationships and calculate an inference. These embeddings can be stored with the supporting IT infrastructure for later re-use. The stored embedding may be utilized at later stages for model fine-tuning or enhancement of context of new input queries.