US 2025/0053300 A1 · EFFECTIVE NOISE REMOVAL TECHNIQUES FOR BIASED MACHINE LEARNING BASED OPTIMIZATIONS IN STORAGE SYSTEMS

Abstract

Techniques for performing effective noise removal for biased machine learning (ML) based optimizations in storage systems. The techniques include serving, by a storage system, an IO workload, identifying, using ML from among a plurality of storage objects subject to the IO workload, storage objects with low temperatures (e.g., cold storage objects) or likely to have low temperatures in the near future, and removing them from subsequent temperature forecasting analysis, effectively treating such cold storage objects as “noise.” The techniques further include performing the temperature forecasting analysis on remaining ones of the plurality of storage objects such as those with high temperatures (e.g., hot storage objects). In this way, temperature forecasting or prediction is performed, using ML, in a biased fashion over a relatively narrow spectrum of storage object temperatures, thereby improving tiering and data prefetching performance, reducing memory and processing overhead, and so on.

Claims20

Dependent claims

A method comprising:

serving an input/output (IO) workload by a storage system, a plurality of storage objects being subject to the IO workload;
identifying, from among the plurality of storage objects, storage objects that may have predetermined temperatures in a near future using a first machine learning (ML) model, the plurality of storage objects including the identified storage objects and a plurality of remaining storage objects;
forecasting, by model inference, temperatures of the plurality of remaining storage objects using a second ML model;
and
performing storage tiering operations on the plurality of remaining storage objects, the storage tiering operations including moving the plurality of remaining storage objects to a specific storage tier of the storage system based on the forecasted temperatures of the plurality of remaining storage objects.

2 more elements

A system comprising:

a memory;
and
processing circuitry configured to execute program instructions out of the memory to:
serve an input/output (IO) workload, a plurality of storage objects being subject to the IO workload;
identify, from among the plurality of storage objects, storage objects that may have predetermined temperatures in a near future using a first machine learning (ML) model, the plurality of storage objects including the identified storage objects and a plurality of remaining storage objects;
forecast, by model inference, temperatures of the plurality of remaining storage objects using a second ML model;
and
perform storage tiering operations on the plurality of remaining storage objects, the storage tiering operations including moving the plurality of remaining storage objects to a specific storage tier of the storage system based on the forecasted temperatures of the plurality of remaining storage objects.

5 more elements

A computer program product including a set of non-transitory, computer-readable media having instructions that, when executed by processing circuitry, cause the processing circuitry to perform a method comprising:

serving an input/output (IO) workload by a storage system, a plurality of storage objects being subject to the IO workload;
identifying, from among the plurality of storage objects, storage objects that may have predetermined temperatures in a near future using a first machine learning (ML) model, the plurality of storage objects including the identified storage objects and a plurality of remaining storage objects;
forecasting, by model inference, temperatures of the plurality of remaining storage objects using a second ML model;
and
performing storage tiering operations on the plurality of remaining storage objects, the storage tiering operations including moving the plurality of remaining storage objects to a specific storage tier of the storage system based on the forecasted temperatures of the plurality of remaining storage objects.

2 more elements

Dependencies are inferred from the claim text.

Effective noise removal techniques for biased machine learning based optimizations in storage systems

Life of this patent

Effective noise removal techniques for biased machine learning based optimizations in storage systems

Life of this patent

Abstract

Claims20