Data Driven Performance Repository to Classify and Retrieve Storage Tuning Profiles

One of the most significant challenges for software defined storage systems is to determine an appropriate tuning configuration considering diverse factors including: hardware, software stacks, and workload characteristics. Currently, the configuration of complex storage systems relies on the experience and skills of human experts. For storage companies, this creates bottlenecks on customer service time, since the number of experts is limited. Moreover, the tuning expertise can be diminished by continuous personal rotation and the lack of experience sharing.
In this presentation, we introduce a data driven framework that continuously integrates and correlates workload characteristics, configuration settings, hardware features, performance logs and benchmarks results, to construct performance tuning profiles. The framework, proposes the use of machine learning techniques to classify profiles based on the description of its components. It also allows the retrieval of most similar tuning solutions to new cases by analyzing the description of the environment.

Learning Objectives:
1. Data driven framework for improving performance of storage systems
2. Architecture of a performance data repository to correlate and blend hardware, software and workload information
3. Machine learning applied to storage information systems