Data repository and method for promoting network storage of data

• US 9,177,175 B2
• Filed: 03/30/2010
• Issued: 11/03/2015
• Est. Priority Date: 02/18/2000
• Status: Expired due to Term

• Current Assignee: Red Hat, Inc. (International Business Machines Corporation)
• Original Assignee: Permabit Technology Corporation (International Business Machines Corporation)
• Inventors: Margolus, Norman H., Floyd, Jered J.
• Primary Examiner(s): Su, Sarah
• Application Number:
  US12/749,603
• Publication Number:
  US 20100185855A1
• Time in Patent Office: 2,044 Days
• Field of Search: 713/165, 713/193, 713/189, 726/26, 726/2, 726/27, 726/30
• US Class Current: 1/1
• CPC Class Codes: G06F 16/137 Hash-based content-based indexing of textual data G06F16/31
• CPC Class Codes: G06F 21/10 Protecting distributed programs or content, e.g. vending or licensing of copyrighted material protection in video systems or pay television H04N7/16 ; Digital rights management [DRM]
• CPC Class Codes: G06F 21/60 Protecting data
• CPC Class Codes: G06F 21/62 Protecting access to data via a platform, e.g. using keys or access control rules
• CPC Class Codes: G06F 21/6209 to a single file or object, e.g. in a secure envelope, encrypted and accessed using a key, or with access control rules appended to the object itself
• CPC Class Codes: G06F 21/6272 by registering files or documents with a third party
• CPC Class Codes: G06F 21/64 Protecting data integrity, e.g. using checksums, certificates or signatures
• CPC Class Codes: G06F 21/645 using a third party
• CPC Class Codes: Y10S 707/959 Network
• CPC Class Codes: Y10S 707/99932 Access augmentation or optimizing
• CPC Class Codes: Y10S 707/99953 Recoverability

Abstract

A method for storing data items in a data repository that allows a client storing a data item to lose the ability to either retrieve the stored data item or enable its retrieval, while preserving controlled access to the stored data item indefinitely. The client storing the data item encrypts it using an encryption key that it creates, which may be derived from the content of the data item. The encryption key is also stored in the data repository in an encrypted form, with its encryption making use of a public key. The method allows storage space to be shared for identical data items encrypted and stored by independent clients of the data repository.

Claims

1. A method for client programs, communicating over a network with a data repository, to store encrypted data items in the data repository, wherein the client programs do not retain information needed to decrypt the data items, the method comprising:

generating a public key and a private key;

transmitting only the public key to a first client program running on a first client machine;

creating a data key, by the first client program;

encrypting a data item, by the first client program, using the data key as the encryption key;

depositing the encrypted data item in the data repository, in response to a request by the first client program;

encrypting the data key, by the first client program, using the public key;

storing the encrypted data key in the data repository in association with the encrypted data item, in response to a request by the first client program;

erasing the data key from storage on the first client machine;

erasing the data item from storage on the first client machine;

retrieving the encrypted data key from the data repository, in response to a request by a second client program running on a second client machine that is different than the first client machine and that has access to the private key;

and

retrieving the encrypted data item and decrypting it;

wherein the private key is needed to decrypt data encoded using the public key;

wherein once both erasing steps have been completed, there is no information stored on the first client machine that would enable the data item to be recovered from the data repository;

wherein the erasing steps guarantee that the data item is no longer accessible using information stored at a site where the first client machine is located;

wherein the retrieving steps occur after the erasing steps have been completed;

wherein the data key erasing step occurs after the encrypted data key is stored in the data repository;

wherein there is no predetermined time limit on when the second client may access the private key;

wherein the storing step associates a repository user with the encrypted data item deposited within the data repository;

wherein the erasing steps guarantee that, using only information present at the site where the first client machine is located and information stored in the data repository, the association of the encrypted data item with the repository user cannot be discovered;

wherein the first client program communicates with the data repository, using hashes of data items to identify their content and determine whether they have content distinct from data items already stored in the data repository, or identical content;

wherein the first client program stores in the data repository data items that have content distinct from data items already stored in the data repository and, for storage of data items that have content identical with data items already stored in the data repository, relies on the data items already stored rather than storing the content again; and

wherein to enable a future retrieval of a data item already stored in the data repository, the first client program communicates to the data repository a proof that it has a data item with identical content and not just a hash used to identify the content, and does this without communicating the full content.

2. The method of claim 1 wherein the depositing and storing steps are part of a backup process that makes copies in the data repository of files stored on the first client machine, allowing recovery from accidental or malicious loss of data on the first client machine without exposing files that have been erased on the first client machine to an agency that gains access to the first client machine.

3. The method of claim 1 wherein the depositing and storing steps are part of an automatic storage process in which, once the erasing steps have been completed, access to the private key is required to recover the data item.

4. The method of claim 3 wherein access to data items stored in the data repository can be given to a third party by giving them access to the data key.

5. The method of claim 4 wherein the ability to read and decrypt the encrypted data item is transmitted from one party to another without retrieving the encrypted data item.

6. The method of claim 1 wherein the data key is a hash of content of the data item.

7. The method of claim 6 wherein the hash used to generate the data key is the same as the hash used by another storage client program storing an identical data item on behalf of a different user of the data repository.

8. The method of claim 6 wherein the hash used to generate the data key is different than the hash used by another storage client program storing an identical data item on behalf of a different user of the data repository.

9. The method of claim 1 wherein a cryptographic hash is used to link a token that identifies the user of the data repository to a collection of data items that have been stored by the first client program operating on the user'"'"'s behalf, and wherein at times when the token is not provided to the repository by a client program, the information stored in the data repository does not allow any of the collection to be identified as belonging to the user.

10. The method of claim 1 wherein the data key is a cryptographic hash of the content of the data item.

11. The method of claim 1 wherein a plurality of identical data items are encrypted by independent client programs using identical data keys derived from the identical data items to produce identical encrypted data items that share storage space in the data repository.

12. The method of claim 1 wherein the data key created for the data item by the first client program is different than the data key that would have been created for the data item by another of the client programs.

13. The method of claim 1 wherein the retrieving steps occur more than a year after the erasing steps have been completed.

14. The method of claim 1 wherein the first client program has no access to the private key or information derived using the private key.

15. The method of claim 1 further comprising:

applying a cryptographic hash to a token that identifies the repository user, and to a handle and version number used by the first client program to identify a repository object owned by the repository user, to create an access identifier that is used by the repository to identify and access the repository object; and

requesting, by the first client program, that the deposited encrypted data item be made part of the repository object;

wherein the encrypted data key is stored as a part of the repository object;

wherein the second client program makes use of the handle and version number and the token in order to access the repository object; and

wherein randomness or pseudorandomness of the version number makes the access identifier random or pseudorandom.

16. The method of claim 15 wherein the handle cannot be determined in a computationally feasible manner using only data stored persistently in the data repository.

17. The method of claim 15 further comprising:

constructing a digital timestamp record by applying a cryptographic hash to the access identifier and to content of the repository object; and

recording the digital timestamp record in a manner that can be used to later cryptographically prove both the time of the recording and the value of the digital timestamp record at the time of the recording,

wherein if the repository object is later modified or deleted, the randomness or pseudorandomness of the access identifier prevents a guess of earlier content of the repository object from being verified using the digital timestamp record.

18. The method of claim 1 wherein a client program relies on the data repository for storage of data items identical in content with data items already stored in the data repository, and stores data items with distinct content elsewhere and not in the data repository.

19. The method of claim 1 wherein the proof is a response to a challenge provided by the data repository, and requires access to at least a portion of the data item with identical content to the data item already stored.

20. The method of claim 1 wherein permission for the future retrieval is recorded by a creation of a named object in the data repository, which links a name and an identified repository user with the data item already stored.