US 12,288,102 B2
System and method for queuing node load in data preprocessing
Mantas Briliauskas, Vilnius (LT)
Assigned to UAB 360 IT, Vilnius (LT)
Filed by UAB 360 IT, Vilnius (LT)
Filed on Mar. 22, 2024, as Appl. No. 18/613,571.
Application 18/613,571 is a continuation of application No. 17/730,775, filed on Apr. 27, 2022, granted, now 11,966,789.
Prior Publication US 2024/0231941 A1, Jul. 11, 2024
Int. Cl. G06F 9/50 (2006.01); G06F 9/48 (2006.01); G06F 21/56 (2013.01)
CPC G06F 9/505 (2013.01) [G06F 9/4881 (2013.01); G06F 21/562 (2013.01); G06F 2221/033 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A computer security system for optimal load distribution and data processing of a plurality of binary files in anti-malware solutions, comprising:
a plurality of node processors each having a copy of an anti-malware machine learning (ML) model; and
a control processor programmed for:
receiving a plurality of binary files used for training of the anti-malware ML models;
separating the plurality of binary files into a plurality of subsets of binary files based on byte size of each of the binary files, and processing workloads amongst all available node processors;
distributing the plurality of subsets of binary files amongst all available node processors to form a distributed work queue by balancing processing workloads across all available node processors such that each node processor processes its respective subset of binary files in parallel and within a similar timeframe as the other node processors,
wherein each of the node processors trains a copy of the anti-malware ML model on a different subset of binary files;
receiving, by the control processor, a report of performance and/or anti-malware processing results of the subset of binary files performed from each node processor; and
updating an anti-malware machine learning algorithm of the machine learning model after all node processors complete processing their workloads by synchronizing model parameters and gradients across all node processors.