US 12,117,975 B2
Linking, deploying, and executing distributed analytics with distributed datasets
Theodoros Salonidis, Wayne, PA (US); Bong Jun Ko, Harrington Park, NJ (US); Swati Rallapalli, Ossining, NY (US); Rahul Urgaonkar, Rye, NY (US); and Dinesh C. Verma, New Castle, NY (US)
Assigned to INTERNATIONAL BUSINESS MACHINES CORPORATION, Armonk, NY (US)
Filed by INTERNATIONAL BUSINESS MACHINES CORPORATION, Armonk, NY (US)
Filed on Sep. 8, 2021, as Appl. No. 17/469,104.
Application 17/469,104 is a continuation of application No. 14/980,733, filed on Dec. 28, 2015, granted, now 11,163,732.
Prior Publication US 2021/0406227 A1, Dec. 30, 2021
This patent is subject to a terminal disclaimer.
Int. Cl. G06F 16/21 (2019.01); G06F 16/23 (2019.01); G06F 16/951 (2019.01); G06F 16/9538 (2019.01); G06F 16/955 (2019.01); G06Q 10/063 (2023.01)
CPC G06F 16/211 (2019.01) [G06F 16/23 (2019.01); G06F 16/951 (2019.01); G06F 16/9538 (2019.01); G06F 16/955 (2019.01); G06Q 10/063 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A computer-implemented method for execution of distributed analytics, comprising:
building a global linked structure that includes correspondences between, and links together, dataset metadata structures that characterize one or more datasets, analytics metadata structures that characterize one or more analytics, and location metadata structures that characterize a plurality of physical computing locations of a distributed computing network and include information defining resources of the plurality of physical computing locations, the global linked structure encoding compatibility between respective datasets, analytics, and locations, including a condition that matches types of data, analytics, and resource constraints of physical location;
determining a set of analytics and a set of compatible datasets compatible with the set of analytics based on the dataset metadata structures, analytics metadata structures, and global linked structure;
determining an optimal execution location from a plurality of physical computing locations in a distributed computing network for execution of the set of analytics on the set of compatible datasets based on costs that are determined based on the location metadata structures, and the global linked structure; and
deploying the sets of analytics and compatible datasets to the optimal execution location.