Granted Patent B2

US 12,566,740 · App. 18/589,255 · Granted Mar 3, 2026

Inventors: Peter Wilczynski (San Francisco, CA); Alexander Ryan (Palo Alto, CA); Allen Chang (Sunnyvale, CA); Daniel Cervelli (Mountain View, CA); Julie Tibshirani (Palo Alto, CA); Timothy Wilson (Palo Alto, CA)

Assignee: Palantir Technologies Inc.

G06F16/219G06F16/2456

Abstract

Claims (69)

1 . A system for analyzing data, comprising:

one or more processors, the processors comprising adapter nodes; and

a memory storing instructions that, when executed by the one or more processors, cause the system to:

receive a request, from an application, for an object;

retrieve, from datastores using a subset of the adapter nodes, data fragments making up or associated with the requested object, wherein:

the adapter nodes are intermediaries between the datastores and application programming interface (API) servers;

transform at least a portion of the data fragments, wherein the transforming comprises:

transforming a first portion of the data fragments based on a first version of a script, the first version of the script being generated according to a first schema of a datastore and a first target dataset ontology of a platform downstream of the application;

receiving an update to the first version of the script, the update resulting in generating a second version of the script, the second version of the script being generated according to the first schema or a second schema of the datastore and the first target dataset ontology or a second target dataset ontology of the platform downstream of the application; and

transforming the first portion or a second portion of the data fragments or modifying the transformed first portion of the data fragments based on the second version of the script;

upon transformation of at least the portion of the data fragments by the subset of the adapter nodes, join the data fragments by the API servers; and

transmit, by the API servers, the joined data fragments to the application.

2 . The system of claim 1 , wherein the adapter nodes further comprise respective caches in which most updated versions of the respective data fragments are stored.

3 . The system of claim 1 , further comprising a data analysis platform that stores a different data fragment; and wherein the instructions further cause the system to:

join the different data fragment to the joined data fragments to form composite data fragments, and wherein the transmitting of the joined data fragment comprises transmitting the composite data fragments.

4 . The system of claim 1 , wherein the instructions further cause the system to:

inactivate an adapter node and activate a new adapter node to replace the inactivated adapter node.

5 . The system of claim 1 , wherein, the instructions further cause the system to:

indicate that the retrieving of the data fragments has been partially completed; and

transmit a partial object view to the application.

6 . The system of claim 1 , wherein the instructions further cause the system to:

receive an indication that a data fragment has updated a version; and

in response to receiving the indication, reingesting, at the application, a new version of the data fragment.

7 . The system of claim 1 , wherein the joining of the data fragments comprises converting a first data fragment of a first data type by a first adapter node and a second data fragment of a second data type by a second adapter node into a standardized data type.

8 . The system of claim 1 , wherein

receive, from the datastores, a new version of a data fragment within a datastore;

in response to receiving the new version of the data fragment, instantiating a new adapter node to retrieve the new version of the data fragment; and

retrieve, from the datastore, using the new adapter node, the new version of the data fragment;

transform at least a portion of the new version of the data fragment based on the first version or the second version of the script; and

upon transformation of at least a portion of the new version of the data fragment, join the new version of the data fragment to other data fragments by the API servers; and

transmit, by the API servers, the joined new version of the data fragment and the other data fragments.

9 . A method being implemented by a computing system including one or more physical processors and storage media storing machine-readable instructions, the method comprising:

receiving a request, from an application, for an object;

retrieving, from datastores using a subset of the adapter nodes, data fragments making up or associated with the requested object, wherein:

the adapter nodes are intermediaries between the datastores and application programming interface (API) servers;

transforming at least a portion of the data fragments, wherein the transforming comprises:

transforming a first portion of the data fragments based on a first version of a script, the first version of the script being generated according to a first schema of a datastore and a first target dataset ontology of a platform downstream of the application;

receiving an update to the first version of the script, the update resulting in generating a second version of the script, the second version of the script being generated according to the first schema or a second schema of the datastore and the first target dataset ontology or a second target dataset ontology of the platform downstream of the application; and

transforming the first portion or a second portion of the data fragments or modifying the transformed first portion of the data fragments based on the second version of the script;

upon transformation of at least the portion of the data fragments by the subset of the adapter nodes, joining the data fragments by the API servers; and

transmitting, by the API servers, the joined data fragments to the application.

10 . The method of claim 9 , further comprising storing most updated versions of the respective data fragments in respective caches of the adapter nodes.

11 . The method of claim 9 , further comprising:

storing a different data fragment in a data analysis platform; and

joining the different data fragment to the joined data fragments to form composite data fragments, wherein the transmitting of the joined data fragment comprises transmitting the composite data fragments.

12 . The method of claim 9 , further comprising:

inactivating an adapter node and activating a new adapter node to replace the inactivated adapter node.

13 . The method of claim 9 , further comprising:

indicating that the retrieving of the data fragments has been partially completed; and

transmitting a partial object view to the application.

14 . The method of claim 9 , wherein a first of the data fragments comprises a tabular format and a second of the data fragments comprises a second, non-object oriented format.

15 . The method of claim 9 , further comprising:

receiving an indication that a data fragment has updated a version; and

in response to receiving the indication, reingesting, at the application, a new version of the data fragment.

16 . A non-transitory computer readable medium comprising instructions that, when executed, cause one or more processors to perform:

receiving a request, from an application, for an object;

retrieving, from datastores using a subset of the adapter nodes, data fragments making up or associated with the requested object, wherein:

the adapter nodes are intermediaries between the datastores and application programming interface (API) servers;

transforming at least a portion of the data fragments, wherein the transforming comprises:

transforming a first portion of the data fragments based on a first version of a script, the first version of the script being generated according to a first schema of a datastore and a first target dataset ontology of a platform downstream of the application;

receiving an update to the first version of the script, the update resulting in generating a second version of the script, the second version of the script being generated according to the first schema or a second schema of the datastore and the first target dataset ontology or a second target dataset ontology of the platform downstream of the application; and

transforming the first portion or a second portion of the data fragments or modifying the transformed first portion of the data fragments based on the second version of the script;

upon transformation of at least the portion of the data fragments by the subset of the adapter nodes, joining the data fragments by the API servers; and

transmitting, by the API servers, the joined data fragments to the application.

17 . The non-transitory computer readable medium of claim 16 , wherein the instructions further cause the one or more processors to perform:

storing most updated versions of the respective data fragments in respective caches of the adapter nodes.

18 . The non-transitory computer readable medium of claim 16 , wherein the instructions further cause the one or more processors to perform:

storing a different data fragment in a data analysis platform; and

joining the different data fragment to the joined data fragments to form composite data fragments, wherein the transmitting of the joined data fragment comprises transmitting the composite data fragments.

Assignments (1)

Continuity (4)

References Cited (94)