Risk assessment of a container build
An artificial intelligence (AI) platform to support selective replacement of one or more image layers of a container image build. A metadata file is subject to natural language processing and one or more corresponding vector representations are created and subject to evaluation by a set of artificial neural networks (ANNs). A first ANN assesses each vector representation with respect to compliance and operability, and the second ANN selectively assesses the vector representation(s) with respect to similarity with one or more compliant vector representations. In response to the assignment of the second score, at least one vector representation of the received metadata file is selectively replaced with at least one compliant vector representation. The metadata file is selectively provisioned with the selectively replaced vector representation(s).
1. A computer system comprising:
a processor operatively coupled to memory;
an artificial intelligence (AI) platform in communication with the processor and memory, the AI platform comprising:
a representation manager configured to employ natural language processing (NLP) to convert a received metadata file representing a container or a virtual machine (VM) image into one or more vector representations;
a neural network manager configured to leverage a first artificial neural network (ANN) to assign a first score to each of the one or more vector representations, the first score to convey a compliance factor corresponding to operability of the one or more vector representations;
the neural network manager configured to selectively leverage a second ANN responsive to the first score assignment from the first ANN, the second ANN configured to assign a second score to the representation of the received metadata file, wherein the second score corresponds to a similarity factor with one or more compliant vector representations;
the neural network manager configured to selectively replace at least one of the vector representations of the received metadata file with at least one of the compliant vector representations responsive to the second score assignment; and
the processor to selectively provision the container or the VM image with the selectively replaced one or more vector representations.
2. The computer system of claim 1 , wherein the first score assignment further comprises the first ANN to measure a distance between a stored vector representation and the one or more vector representations of the received metadata file, and wherein the first score assignment is based on the measured distance.
3. The computer system of claim 2 , further comprising responsive to the first score assignment, the second ANN configured to identify a stored compliant vector representation closest to the vector representation associated with the first score, and measure the distance between the identified stored compliant vector representation and the vector representation associated with the first score.
4. The computer system of claim 1 , further comprising the representation manager to filter the received metadata file and remove noise from the metadata file prior to the conversion of the metadata file to the one or more vector representations.
5. The computer system of claim 4 , wherein the filtered metadata file is comprised of container or VM layers of executable code.
6. The computer system of claim 5 , wherein selectively replacing at least one of the vector representations further comprises the neural network manager to individually process remaining container or VM layers of executable code for compliance, and selectively and individually swap one or more of the remaining container or VM layers of executable code responsive to the individual processing.
7. A computer program product comprising:
a computer readable storage device; and
program code embodied with the computer readable storage device, the program code executable by a processor to:
employ natural language processing (NLP) to convert a received metadata file representing a container or a virtual machine (VM) image into one or more vector representations;
leverage a first artificial neural network (ANN) to assign a first score to each of the one or more vector representations, the first score configured to convey a compliance factor corresponding to operability;
leverage a second ANN responsive to the first score assignment from the first ANN, the second ANN configured to assign a second score to a selective set of the one or more vector representations, the second score configured to convey a similarity factor with one or more compliant vector representations;
selectively replace at least one of the vector representations of the selective set with at least one of the compliant vector representations responsive to the second score assignment; and
provision the container or the VM image containing the selectively replaced at least one vector representation.
8. The computer program product of claim 7 , wherein assignment of the first score further comprises program code configured to leverage the first AAN to measure a distance between a stored vector representation and the one or more vector representations subject to the compliance factor conveyance, and wherein the first score assignment is based on the measured distance.
9. The computer program product of claim 8 , further comprising responsive to the first score assignment, the second ANN configured to identify a stored compliant vector representation closest to the vector representation associated with the first score, and measure the distance between the identified stored compliant vector representation and the vector representation associated with the first score.
10. The computer program product of claim 7 , further comprising program code configured to filter the received metadata file and remove noise from the metadata file prior to the conversion of the metadata file to the one or more vector representations.
11. The computer program product of claim 10 , wherein the filtered metadata file is comprised of container or VM layers of executable code.
12. The computer program product of claim 11 , wherein selectively replacing at least one of the vector representations further comprises program code configured to individually process remaining container or VM layers of executable code for compliance, and selectively and individually swap one or more of the remaining container or VM layers of executable code responsive to the processing.
13. A computer-implemented method comprising:
employing natural language processing (NLP), converting a received metadata file representing a container or a virtual machine (VM) image into one or more vector representations;
leveraging a first artificial neural network (ANN) to assign a first score to each of the one or more vector representations, the first score conveying a compliance factor corresponding to operability;
selectively leveraging a second ANN responsive to the first score assignment from the first ANN, the second ANN assigning a second score to each of the one or more vector representations, the second score conveying a similarity factor with one or more compliant vector representations; and
selectively replacing at least one of the vector representations of the received metadata file with at least one of the compliant vector representations responsive to the second score assignment, and provisioning the container or the VM image received metadata file containing the selectively replaced at least one vector representation.
14. The computer-implemented method of claim 13 , wherein assigning the first score further comprises the first ANN measuring a distance between a stored vector representation and the one or more vector representations subject to the provisioning, and wherein the first score assignment is based on the measured distance.
15. The computer-implemented method of claim 14 , further comprising responsive to the first score assignment, the second ANN identifying a stored compliant vector representation closest to the vector representation associated with the first score, and measuring the distance between the identified stored compliant vector representation and the vector representation associated with the first score.
16. The computer-implemented method of claim 13 , further comprising filtering the received metadata file and removing noise from the metadata file prior to the conversion of the metadata file to the one or more vector representations.
17. The computer-implemented method of claim 16 , wherein the filtered metadata file is comprised of container or VM layers of executable code.
18. The computer-implemented method of claim 17 , wherein selectively replacing at least one of the vector representations further comprises individually processing remaining container or VM layers of executable code for compliance, and selectively and individually swapping one or more of the remaining container or VM layers of executable code responsive to the processing.