US 12,430,169 B2
Storage system with improved quality of service
Byeong Hui Kim, Hwaseong-si (KR); Dong Hyub Kang, Seoul (KR); Hyun Kyo Oh, Yongin-si (KR); Sung Min Jang, Hwaseong-si (KR); and Ki Been Jung, Incheon (KR)
Assigned to SAMSUNG ELECTRONICS CO., LTD., Suwon-si (KR)
Filed by SAMSUNG ELECTRONICS CO., LTD., Suwon-si (KR)
Filed on Aug. 11, 2022, as Appl. No. 17/819,007.
Claims priority of application No. 10-2021-0161470 (KR), filed on Nov. 22, 2021.
Prior Publication US 2023/0161619 A1, May 25, 2023
Int. Cl. G06F 9/48 (2006.01); G06N 5/022 (2023.01)
CPC G06F 9/4881 (2013.01) [G06N 5/022 (2013.01)] 17 Claims
OG exemplary drawing
 
1. A storage system comprising:
a storage device comprising a plurality of non-volatile memory devices; and
a storage controller configured to access the storage device, and to perform a workload based on a parameter,
wherein the parameter is selected by a simulation model to increase a Quality of Service (QOS) for the workload,
wherein the simulation model is contained on a machine-learning model embedded device connected to the storage system and configured to:
perform a training workload using a training parameter to generate predicted QoS data based on an initial simulation model, wherein the storage controller is also configured to perform the training workload using the training parameter to calculate real QoS data;
calculate a loss based on a comparison between the predicted QOS data and the real QoS data;
train the initial simulation model to update the simulation model based on the loss;
cluster the training parameter and the training workload based on the real QOS data to find features;
label the training parameter and the training workload based on the features; and
operate based on the labeled training parameter and the labeled training workload to generate the predicted QoS data.