Anomaly and ransomware detection
Some examples relate generally to computer architecture software for information security and, in some more particular aspects, to machine learning based on changes in snapshot metadata for anomaly and ransomware detection in a file system.
1. A system, comprising:
a storage device configured to store one or more snapshots of a primary machine; and
one or more processors in communication with the storage device and a production system, the one or more processors configured to perform anomaly and ransomware detection operations, comprising:
taking a first snapshot and a second snapshot of the primary machine and storing the first snapshot and the second snapshot in the storage device;
generating a first differential filesystem metadata (diff FMD) file by sampling from a first set of metadata files of a first seed dataset that simulates normal operation of the primary machine, and a second diff FMD file by sampling from a second set of metadata files of a second seed dataset that simulates a ransomware infection, the first diff FMD file and the second diff FMD file being indicative of one or more changes of at least one file of the primary machine occurring between the first snapshot and the second snapshot, the one or more changes of the at least one file being represented by respective changes in snapshot-based filesystem metadata of the first diff FMD file and the second diff FMD file;
generating snapshot-based training data by merging the first diff FMD file and the second diff FMD file, wherein the merging of the first diff FMD file and the second diff FMD file simulates ransomware activity for live production data using the snapshot-based filesystem metadata, and wherein the simulated ransomware activity is based on the one or more changes in the snapshot-based filesystem metadata of the first diff FMD file and the second diff FMD file and based on the one or more changes of the at least one file of the primary machine occurring between the first snapshot and the second snapshot;
training one or more machine-learning models using the snapshot-based training data; and
generating an anomaly prediction by inputting a third diff FMD file from a third snapshot into a first machine-learning model of the one or more machine-learning models trained using the snapshot-based training data and an encryption prediction by inputting a result of the anomaly prediction of the first machine-learning model into a second machine-learning model of the one or more machine-learning models to determine whether the anomaly prediction is a ransomware encryption anomaly.
2. The system of claim 1 , wherein the storage device and the one or more changes in the snapshot-based filesystem metadata are received from a backup system that includes a backup storage device.
3. The system of claim 1 , wherein the anomaly and ransomware detection operations are performed without impacting the production system.
4. The system of claim 3 , wherein one or more of the anomaly and ransomware detection operations are offloaded to a cloud-based software-as-a-service platform.
5. The system of claim 1 , wherein the first machine-learning model of the one or more machine-learning models includes an anomaly model, and the second machine-learning model of the one or more machine-learning models includes an encryption model.
6. The system of claim 1 , wherein training the one or more machine-learning models is based on the snapshot-based training data, the snapshot-based training data being derived from snapshot-based metadata.
7. A system, comprising:
one or more processors in communication with a storage device and a production system, the one or more processors configured to perform a computer implemented method including anomaly and ransomware detection operations, comprising:
taking a first snapshot and a second snapshot of a primary machine and storing the first snapshot and the second snapshot in the storage device;
generating a first differential filesystem metadata (diff FMD) file by sampling from a first set of metadata files of a first seed dataset that simulates normal operation of the primary machine, and a second diff FMD file by sampling from a second set of metadata files of a second seed dataset that simulates a ransomware infection, the first diff FMD file and the second diff FMD file being indicative of one or more changes of at least one file of the primary machine occurring between the first snapshot and the second snapshot, the one or more changes of the at least one file being represented by respective changes in snapshot-based filesystem metadata of the first diff FMD file and the second diff FMD file;
generating snapshot-based training data by merging the first diff FMD file and the second diff FMD file, wherein the merging of the first diff FMD file and the second diff FMD file simulates ransomware activity for live production data using the snapshot-based filesystem metadata, and wherein the simulated ransomware activity is based on the one or more changes in the snapshot-based filesystem metadata of the first diff FMD file and the second diff FMD file and based on the one or more changes of the at least one file of the primary machine occurring between the first snapshot and the second snapshot;
training one or more machine-learning models using the snapshot-based training data; and
generating an anomaly prediction by inputting a third diff FMD file from a third snapshot into a first machine-learning model of the one or more machine-learning models trained using the snapshot-based training data and an encryption prediction by inputting a result of the anomaly prediction of the first machine-learning model into a second machine-learning model of the one or more machine-learning models to determine whether the anomaly prediction is a ransomware encryption anomaly.
8. The system of claim 7 , wherein the storage device and the one or more changes in the snapshot-based filesystem metadata are received from a backup system that includes a backup storage device.
9. The system of claim 7 , wherein the anomaly and ransomware detection operations are performed without impacting the production system.
10. The system of claim 9 , wherein one or more of the anomaly and ransomware detection operations are offloaded to a cloud-based software-as-a-service platform.
11. The system of claim 7 , wherein the first machine-learning model of the one or more machine-learning models includes an anomaly model, and the second machine-learning model of the one or more machine-learning models includes an encryption model.
12. The system of claim 7 , wherein training the one or more machine-learning models is based on the snapshot-based training data, the snapshot-based training data being derived from snapshot-based metadata.
13. A non-transitory, machine-readable medium, comprising:
instructions which, when read by a machine, cause the machine to perform anomaly and ransomware detection operations comprising:
taking a first snapshot and a second snapshot of a primary machine and storing the first snapshot and the second snapshot in a storage device;
generating a first differential filesystem metadata (diff FMD) file by sampling from a first set of metadata files of a first seed dataset that simulates normal operation of the primary machine, and a second diff FMD file by sampling from a second set of metadata files of a second seed dataset that simulates a ransomware infection, the first diff FMD file and the second diff FMD file being indicative of one or more changes of at least one file of the primary machine occurring between the first snapshot and the second snapshot, the one or more changes of the at least one file being represented by respective changes in snapshot-based filesystem metadata of the first diff FMD file and the second diff FMD file;
generating snapshot-based training data by merging the first diff FMD file and the second diff FMD file, wherein the merging of the first diff FMD file and the second diff FMD file simulates ransomware activity for live production data using the snapshot-based filesystem metadata, and wherein the simulated ransomware activity is based on the one or more changes in the snapshot-based filesystem metadata of the first diff FMD file and the second diff FMD file and based on the one or more changes of the at least one file of the primary machine occurring between the first snapshot and the second snapshot;
training one or more machine-learning models using the snapshot-based training data; and
generating an anomaly prediction by inputting a third diff FMD file from a third snapshot into a first machine-learning model of the one or more machine-learning models trained using the snapshot-based training data and an encryption prediction by inputting a result of the anomaly prediction of the first machine-learning model into a second machine-learning model of the one or more machine-learning models to determine whether the anomaly prediction is a ransomware encryption anomaly.
14. The non-transitory, machine-readable medium of claim 13 , wherein the storage device and the one or more changes in the snapshot-based filesystem metadata are received from a backup system that includes a backup storage device.
15. The non-transitory, machine-readable medium of claim 13 , wherein the anomaly and ransomware detection operations are performed without impacting a production system.
16. The non-transitory, machine-readable medium of claim 15 , wherein one or more of the anomaly and ransomware detection operations are offloaded to a cloud-based software-as-a-service platform.
17. The non-transitory, machine-readable medium of claim 13 , wherein the first machine-learning model of the one or more machine-learning models includes an anomaly model, and the second machine-learning model of the one or more machine-learning models includes an encryption model.
18. The non-transitory, machine-readable medium of claim 13 , wherein training of the one or more machine-learning models is based on the snapshot-based training data, the snapshot-based training data being derived from snapshot-based metadata.