US 12,407,577 B2
Graph neural network based cloud traffic prediction and optimization
Lei Xu, Shanghai (CN); and Deng Feng Wan, Shanghai (CN)
Assigned to SAP SE, Walldorf (DE)
Filed by SAP SE, Walldorf (DE)
Filed on May 11, 2022, as Appl. No. 17/741,795.
Prior Publication US 2023/0370337 A1, Nov. 16, 2023
Int. Cl. H04L 41/147 (2022.01); G06N 3/044 (2023.01); G06N 3/08 (2023.01); H04L 67/02 (2022.01)
CPC H04L 41/147 (2013.01) [G06N 3/044 (2023.01); G06N 3/08 (2013.01); H04L 67/02 (2013.01)] 21 Claims
OG exemplary drawing
 
1. A system to facilitate traffic prediction for a cloud computing environment, comprising:
a traffic prediction server, including:
a computer processor, and
a memory storage device, coupled to the computer processor, including instructions that, when executed by the computer processor, enable the traffic prediction server to:
(i) retrieve performance stack trace logs from a traffic performance stack trace log repository,
(ii) parse the performance stack trace logs as an objects list including parent/child object relationships,
(iii) store the parsed objects list in a graph database,
(iv) transform the graph database into training data including spatial and temporal information, and
(v) use the transformed training data to train a transformer model.