Encrypted search indexes
In one example, a method for searching data includes creating a set of tokens corresponding to data of an encrypted dataset, and then encrypting the tokens. Next, an encrypted search index is created that includes the encrypted tokens and that excludes any positional information concerning the data to which the encrypted tokens correspond. A query is then defined that includes one or more encrypted search tokens. Next, search results are received that identify any matches between the encrypted search tokens and the encrypted tokens of the encrypted search index. Finally, the data that corresponds to any identified tokens that match a token of the encrypted search index is decrypted.
1. A method for searching data, comprising:
creating a set of tokens corresponding to data of an encrypted dataset;
encrypting the tokens;
creating an encrypted search index that includes the encrypted tokens and that excludes any positional information concerning the data to which the encrypted tokens correspond;
defining a query that includes one or more encrypted search tokens;
receiving search results that identify any matches between the encrypted search tokens and the encrypted tokens of the encrypted search index; and
decrypting the data that corresponds to any identified tokens that match a token of the encrypted search index.
2. The method as recited in claim 1 , wherein one or more of the tokens created based on the dataset correspond to multiple words of the dataset.
3. The method as recited in claim 1 , wherein the dataset is a backup saveset.
4. The method as recited in claim 1 , wherein the encrypted search index and the encrypted tokens of the data set are stored at a location remote from a location where the encrypted search index was created.
5. The method as recited in claim 1 , wherein the encrypted tokens are listed in the encrypted search index in either alphabetical order or random order.
6. The method as recited in claim 1 , wherein a first one of the tokens is a split token that is associated with a plurality of other tokens that each include a portion of the data with which the first token is associated.
7. The method as recited in claim 1 , wherein a first one of the tokens is a combined token that includes the respective data with which two or more other tokens are associated.
8. The method as recited in claim 1 , wherein the query identifies an unquoted quoted string.
9. The method as recited in claim 1 , wherein the query identifies a quoted string.
10. The method as recited in claim 9 , wherein none of the matches found exactly match the quoted string.
11. The method as recited in claim 1 , wherein the query specifies a single word token, or a multi-word token.
12. A non-transitory storage medium having stored therein computer-executable instructions which, when executed by one or more hardware processors, implement the processes recited in claim 1 .
13. A client, comprising:
one or more hardware processors; and
the non-transitory storage medium as recited in claim 12 .
14. A non-transitory storage medium having stored therein computer-executable instructions which, when executed by one or more hardware processors, perform the following:
creating a set of tokens corresponding to data of an encrypted dataset;
encrypting the tokens;
creating an encrypted search index that includes the encrypted tokens and that excludes any positional information concerning the data to which the encrypted tokens correspond;
defining a query that includes one or more encrypted search tokens;
receiving search results that identify any matches between the encrypted search tokens and the encrypted tokens of the encrypted search index; and
decrypting the data that corresponds to any identified tokens that match a token of the encrypted search index.
15. A client, comprising:
one or more hardware processors; and
the non-transitory storage medium as recited in claim 14 .
16. A method for searching data, comprising:
receiving a set of encrypted tokens that correspond to a backup dataset;
storing the set of encrypted tokens;
receiving an encrypted search index that lists the encrypted tokens;
storing the encrypted search index in association with the set of encrypted tokens;
receiving a query that identifies one or more search tokens; and
searching the set of encrypted tokens for the one or more search tokens.
17. The method as recited in claim 16 , wherein the backed up dataset is encrypted.
18. The method as recited in claim 16 , further comprising identifying any matches between the search tokens and the encrypted tokens listed in the encrypted search index.
19. A non-transitory storage medium having stored therein computer-executable instructions which, when executed by one or more hardware processors, implement the processes recited in claim 16 .
20. A client, comprising:
one or more hardware processors; and
the non-transitory storage medium as recited in claim 19 .