Preventive hash loading
First Claim
1. A method for preventive hash loading, the method comprising:
- receiving an indication at a storage server that a machine will be backed up, wherein the storage server is connected to the one or more machines by one or more networks;
creating a fingerprint cache in RAM using a backup agent;
loading fingerprints of blocks related to a previous backup of the machine to the fingerprint cache of the storage server;
searching the storage server for fingerprints in the fingerprint cache that match fingerprints of incoming blocks from the machine being backed up;
in response to determining that the fingerprints of the incoming blocks do not match fingerprints in the fingerprint cache, searching for the fingerprints in a database; and
increasing speed of backing up the machine by transferring only blocks from the machine being backed up that are not in the fingerprint cache or the database of the storage server to the storage server, wherein searching fingerprint cache is faster than searching database.
5 Assignments
0 Petitions
Accused Products
Abstract
A method, computer program product, computing system, and system for preventive hash loading are described. The method may include receiving an indication at a storage server that a machine will be backed up. The method may further include loading fingerprints of blocks related to a previous backup of the machine to RAM of the storage server. The method may also include searching the storage server for fingerprints in the RAM that match fingerprints of incoming blocks from the machine being backed up. The method may additionally include, in response to determining that the fingerprints of the incoming blocks do not match fingerprints in the RAM, searching for the fingerprints in a database. Moreover, the method may include transferring only blocks from the machine being backed up that are not in the RAM or the database of the storage server to the storage server.
-
Citations
16 Claims
-
1. A method for preventive hash loading, the method comprising:
-
receiving an indication at a storage server that a machine will be backed up, wherein the storage server is connected to the one or more machines by one or more networks; creating a fingerprint cache in RAM using a backup agent; loading fingerprints of blocks related to a previous backup of the machine to the fingerprint cache of the storage server; searching the storage server for fingerprints in the fingerprint cache that match fingerprints of incoming blocks from the machine being backed up; in response to determining that the fingerprints of the incoming blocks do not match fingerprints in the fingerprint cache, searching for the fingerprints in a database; and increasing speed of backing up the machine by transferring only blocks from the machine being backed up that are not in the fingerprint cache or the database of the storage server to the storage server, wherein searching fingerprint cache is faster than searching database. - View Dependent Claims (2, 3, 4, 5)
-
-
6. A computer program product residing on a computer readable storage medium having a plurality of instructions stored thereon, which, when executed by a processor, cause the processor to perform operations for preventive hash loading, the operations comprising:
-
receiving an indication at a storage server that a machine will be backed up; creating a fingerprint cache in RAM of the storage server; loading fingerprints of blocks related to a previous backup of the machine to the fingerprint cache; searching the storage server for fingerprints in the fingerprint cache that match fingerprints of incoming blocks from the machine being backed up; in response to determining that the fingerprints of the incoming blocks do not match fingerprints in the fingerprint cache, searching for the fingerprints in a database; increasing speed of backing up the machine by transferring only blocks from the machine being backed up that are not in the fingerprint cache or the database of the storage server to the storage server, wherein searching fingerprint cache is faster than searching database; and replenishing fingerprint cache depending on the backed up data blocks. - View Dependent Claims (7, 8, 9, 10)
-
-
11. A computing system for preventive hash loading, the computing system comprising one or more processors, wherein the one or more processors are configured to:
-
receive an indication at a storage server that a machine will be backed up; create a fingerprint cache in RAM using a backup agent; load fingerprints of blocks related to a previous backup of the machine to the fingerprint cache of the storage server; search the storage server for fingerprints in the fingerprint cache that match fingerprints of incoming blocks from the machine being backed up; in response to determining that the fingerprints of the incoming blocks do not match fingerprints in the fingerprint cache, search for the fingerprints in a database; and increase speed of backing up the machine by transferring only blocks from the machine being backed up that are not in the fingerprint cache RAM or the database of the storage server to the storage server, wherein searching fingerprint cache is faster than searching database. - View Dependent Claims (12, 13, 14, 15)
-
-
16. A system for preventive hash loading, the system comprising:
-
a backup agent configured to perform backups of one or more machines; a storage server that received the backups, wherein the storage server is connected to the one or more machines by one or more networks; a database for storing fingerprints of blocks received during the backups; and a RAM in the storage server; and a fingerprint cache created in the RAM, the finger print cache configured to load fingerprints of blocks corresponding to a previous backup of a machine being backed up, wherein fingerprint cache is replenished depending on the backed up data blocks, wherein fingerprint cache stores hashes of recently backed up blocks and hashes of blocks that are frequently used, wherein backup agent increases speed of backing up the machine by transferring only blocks from the machine being backed up that are not in the RAM or the database of the storage server to the storage server such that hash traffic through network is reduced.
-
Specification