IBM Assigned Eleven Patents

IBM Assigned Ten Patents
Virtualization, de-dupe, data allocation, flash systems, SSD and hybrid drives

Virtualizing storage systems
and managing data independently
IBM Corp., Armonk, NY, has been assigned a patent (8,423,737) developed by Nils Haustein, Soergenloch, Germany, and Thorsten Krause, Mainz, Germany, for "systems and methods for virtualizing storage systems and managing data independently."

The abstract of the patent published by the U.S. Patent and Trademark Office states: "Method, data processing systems, and computer program products are provided for virtualizing and managing a storage virtualization system in a storage management architecture. Source data is copied from the source storage media to target data in a target storage media based on a predefined copy policy in a copy mapping table. A relation between the source data and the target data is tracked in a copy mapping table. It is determined if a copy of the requested data exists using the copy mapping table. A least utilized storage system having a copy of the requested storage media is determined. Access to the requested storage media in the least utilized storage system is tested. If access is not possible, access to a copy of the requested storage media in another storage system is provided by updating a frontend-backend mapping table and forwarding all data access commands to the other system."

The patent application was filed on Oct. 21, 2010 (12/909,764).

Data allocation in distributed storage system
IBM Corp., Armonk, NY, has been assigned a patent (8,452,899) developed by four co-inventors for a "data allocation in a distributed storage system."

The co-inventors are Ofir Zohar, Alfei Menashe, Israel, Yaron Revah, Tel-Aviv, Israel, Haim Helman, Ramat Gan, Israel, and Dror Cohen, Petach-Tikva, Israel.

The abstract of the patent published by the U.S. Patent and Trademark Office states: "A method for data distribution, including distributing logical addresses among an initial set of devices so as provide balanced access, and transferring the data to the devices in accordance with the logical addresses. If a device is added to the initial set, forming an extended set, the logical addresses are redistributed among the extended set so as to cause some logical addresses to be transferred from the devices in the initial set to the additional device. There is substantially no transfer of the logical addresses among the initial set. If a surplus device is removed from the initial set, forming a depleted set, the logical addresses of the surplus device are redistributed among the depleted set. There is substantially no transfer of the logical addresses among the depleted set. In both cases the balanced access is maintained."

The patent application was filed on December 15, 2011.
 
De-dupe in network storage
with data residence compliance
IBM Corp.,  Armonk, NY, has been assigned a patent (8,447,732) developed by four co-inventors for a "deduplication in network storage with data residence compliance."

The co-inventors are Abhinay R. Nagpal, Maharashtra, India, Sandeep R. Patil, Elmsford, NY, Gandhi Sivakumar, Melbourne, Australia, and Carolyn A. Whitehead, Norwich, CT.

The abstract of the patent published by the U.S. Patent and Trademark Office states: "Deduplication in a network storage environment includes, for files stored in a network, determining a location constraint status specified by a compliance agreement for each of the files. Location constraint statuses include a location of persistent residency and no residency restriction. Deduplication also includes selecting a file from the files in the network and identifying corresponding redundant files, the selected file and the corresponding redundant files representing a set. Deduplication further includes determining the location constraint status for each of the files in the set. For the files in the set having a location constraint status specifying a location of persistent residency, the deduplication includes retaining a master copy at the respective location of persistent residency, and removing the corresponding redundant files from the network."

The patent application was filed on Aug. 2, 2011 (13/196,285).

Reducing current draw of a plurality of SSDs
at computer start-up
IBM Corp., Armonk, NY, has been assigned a patent (8,447,998) developed by Mark E. Andresen, and Joaquin F. Pacheco, Cary, NC, for "reducing current draw of a plurality of solid state drives at computer start-up."

The abstract of the patent published by the U.S. Patent and Trademark Office states: "Reducing current draw of solid state drives from a shared power supply of a computer at computer start-up, each SSD including computer memory, a capacitor, a disk controller, and a charge controller, the disk controller configured to enable the charge controller to charge the capacitor upon receiving a charge command, the SSDs organized into start-up groups characterized by a position in a predefined start-up order. Upon start-up of the computer, beginning with a first start-up group in the predefined start-up order and until the last start-up group in the predefined start-up order has received a charge command, embodiments include, sending, by a storage device initiator, a charge command to a start-up group to initiate charging of the capacitor of each solid state drive in the start-up group and waiting a predefined period of time before sending another charge command to a next start-up group in the predefined start-up order."

The patent application was filed on May 22, 2012 (13/477,257). The

Failure-specific data collection and recovery
for enterprise storage controllers
IBM Corp., Armonk, NY, has been assigned a patent (8,448,013) developed by seven co-inventors for a "failure-specific data collection and recovery for enterprise storage controllers."

The co-inventors are Brian Dow Clark, Tucson, AZ, Juan Alonso Coronado, Tucson, AZ, Christina Ann Lara, Lisa R. Martinez, Phu Nguyen, Beth Ann Peterson, and Jayson Elliott Tsingine, Tucson, AZ.

The abstract of the patent published by the U.S. Patent and Trademark Office states: "A method, apparatus, and computer program product for handling a failure condition in a storage controller is disclosed. In certain embodiments, a method may include initially detecting a failure condition in a storage controller. The failure condition may be associated with a specific host and a specific storage device connected to the storage controller. The method may further include determining a failure ID associated with the failure condition. Using the failure ID, en entry may be located in a data collection and recovery table. This entry may indicate one or more data collection and/or recovery processes to execute in response to the failure condition. The method may then execute the data collection and/or recovery processes indicated in the entry. While executing the data collection and/or recovery processes, connectivity may be maintained between hosts and storage devices not associated with the failure condition."

The patent application was filed on Jan. 19, 2009 (12/355,944).

Protecting data using variable size page stripes
in flash-based storage system
IBM Corp., Armonk, NY, has been assigned a patent (8,443,136) developed by Holloway H. Frost, Houston, James A. Fuxa, Houston, and Charles J. Camp, Sugar Land, TX, for a "method and apparatus for protecting data using variable size page stripes in a flash-based storage system."

The abstract of the patent published by the U.S. Patent and Trademark Office states: "Methods and apparatuses for enhanced protection of data stored in a flash memory system involve a controller capable of using variable size page stripes in the memory system. The controller is configured to store data such that each page stripe comprises a plurality of data pages, with each data page in the page stripe being stored in a different flash memory chip. The controller is also configured to maintain one or more buffers containing information reflecting blocks of memory within the flash memory chips that have been erased and are available for information storage, and to dynamically determine the number of data pages to be included in a page stripe based on the information in the one or more buffers such that a first page stripe and a second page stripe can have different numbers of data pages."

The patent application was filed on Dec. 17, 2010 (12/971,286).

Generating accurate performance targets
for active semiconductor devices
IBM Corp., Armonk, NY, has been assigned a patent (8,453,101) developed by four co-inventors for a "method, system and program storage device for generating accurate performance targets for active semiconductor devices during new technology node development."

The co-inventors are James M. Johnson, Milton, VT, Scott K. Springer, Burlington, VT, Rainer Thoma, Essex Junction, VT, and Josef S. Watts, South Burlington, VT.

The abstract of the patent published by the U.S. Patent and Trademark Office states: "Disclosed are embodiments of a method, system and program storage device for generating accurate performance targets for active semiconductor devices during technology node development in order to reduce the number of iterations required for model extraction and/or to improve model quality. In these embodiments, initial sets of performance targets for related semiconductor devices are generated, e.g., by making assumptions based on hardware measurements taken from semiconductor devices in prior technology nodes. Additional processes are then performed on the initial sets of performance targets prior to the modeling stage in order to detect and resolve any inconsistencies between the data in the sets. Specifically, plotting techniques are performed with respect to the performance targets. The results are analyzed to detect any inconsistencies indicating that the performance targets are inaccurate and adjustments are made to the performance targets in order to resolve those inconsistencies."

The patent application was filed on Nov. 22, 2011 (13/302,350).

Securing distributed de-dupe software
IBM Corp., Armonk, NY, has been assigned a patent (8,453,257) developed by six co-inventors for an "approach for securing distributed deduplication software."

The co-inventors are Matthew J. Anglin, Tucson, AZ, David M. Cannon, Tucson, AZ, Avishai H. Hochberg, San Jose, CA, Alexei Kojenov, Beaverton, OR, James P. Smith, Redwood City, CA, and Mark L. Yakushev, San Jose, CA.

The abstract of the patent published by the U.S. Patent and Trademark Office states: "The various embodiments of the present invention include techniques for securing the use of data deduplication activities occurring in a source-deduplicating storage management system. These techniques are intended to prevent fake backup, target data contamination, and data spoofing attacks initiated by a source. In one embodiment, one technique includes limiting chunk querying to authorized users. Another technique provides detection of attacks and unauthorized access to keys within the target system. Additional techniques include the combination of validating the existence of data from the source by validating the data chunk, validating a data sample of the data chunk, or validating a hash value of the data chunk. A further embodiment involves the use of policies to provide authorization levels for chunk sharing and linking within the target. These techniques separately and in combination provide a comprehensive strategy to avoid unauthorized access to data within the target storage system."

The patent application was filed on Aug. 14, 2009 (12/541,191).

HDD using non-volatile storage device
as cache for modified tracks
IBM Corp., Armonk, NY, has been assigned a patent (8,432,632) developed by Michael T. Benhase and Lokesh M. Gupta, Tucson, AZ, for a "magnetic disk drive using a non-volatile storage device as cache for modified tracks."

The abstract of the patent published by the U.S. Patent and Trademark Office states: "Provided are a computer program product, system, and method for a magnetic disk drive. The disk drive has at least one disk platter having at least one recordable disk surface having an areal density of at least 200 gigabits per square inch. Either a diameter of the at least one disk platter is greater than 3.5 inches or the at least one disk platter rotates at less than 5400 RPMs. A R/W head reads and writes tracks of data with respect to the at least one disk surface. Modified tracks from write requests to write to the at least one disk surface on the at least one disk platter are cached in a non-volatile storage device for caching modified tracks. Modified tracks are cached in the non-volatile storage device to later destage to the at least one disk surface."

The patent application was filed on May 23, 2011 (13/113,958).

Hybrid storage subsystem
with mixed placement of file contents
IBM Corp., Armonk, NY, has been assigned a patent (8,438,334) developed by four co-inventors for a "hybrid storage subsystem with mixed placement of file contents."

The co-inventors are Robert H. Bell Jr., Austin, TX, Men-Chow Chiang, Hong L. Hua, and Mysore S. Srinivas, Austin, TX.

The abstract of the patent published by the U.S. Patent and Trademark Office states: "A storage subsystem combining SSD and HDD technologies provides low access latency and low complexity. Separate free lists are maintained for the SSD and the HDD and blocks of file system data are stored uniquely on either the SSD or the HDD. When a read access is made to the subsystem, if the data is present on the SSD, the data is returned, but if the block is present on the HDD, it is migrated to the SSD and the block on the HDD is returned to the HDD free list. On a write access, if the block is present in the either the SSD or HDD, the block is overwritten, but if the block is not present in the subsystem, the block is written to the HDD."

The patent application was filed on Dec. 22, 2009 (12/644,721).


Saving log data using HDD system as primary cache
and tape library as secondary cache
IBM Corp., Armonk, NY, has been assigned a patent (8,458,397) developed by six co-inventors for "saving log data using a disk system as primary cache and a tape library as secondary cache."

The co-inventors are Norie Iwasaki, Fujisawa, Japan, Koichi Masuda, Yokohama, Japan, Tadaaki Minoura, Yamato, Japan, Tomokazu Nakamura, Kawasaki, Japan, Takeshi Sohda, Sagamihara, Japan, and Takahiro Tsuda, Yokohama, Japan.

The abstract of the patent published by the U.S. Patent and Trademark Office states: "Various embodiments save a plurality of log data in a HSM system using a disk system as a primary cache with a tape library as a secondary cache. The user data is stored in the primary cache and written into the secondary cache at a subsequent period of time. The plurality of blank tapes in the secondary cache is prepared for storing the user data and the plurality of log data based on priorities. At least one of the plurality of blank tapes is selected for copying the plurality of log data and the user data from the primary cache to the secondary cache based on priorities. The plurality of log data is stored in the primary cache. The selection of at least one of the plurality of blank tapes completely filled with the plurality of log data is delayed for writing additional amounts of user data."

The patent application was filed on June 11, 2012 (13/493,180).

Data backup
IBM Corp., Armonk, NY, has been assigned a patent (8,464,010) developed by four co-inventors for an "apparatus and method for data backup."

The co-inventors are Christopher B. E. Beeken, Eastleigh, UK, Carlos F. Fuente, Portsmouth, UK, Colin R. Jewell, Chandlers Ford, UK, and William J. Scales, Portchester, UK.

The abstract of the patent published by the U.S. Patent and Trademark Office states: "A targeting storage component selects a target virtual disk for one of a full copy or an incremental copy of first data. A periodic backup component triggers a periodic point-in-time copy of the first data to a virtual disk in the cascade. A testing component tests a status of the full copy, the incremental copy and the periodic point-in-time copy. Responsive to the status, a cascade splitting component splits the cascade to remove a dependency relationship of at least one of the fully copy, the incremental copy and the periodic point-in-time copy on the first data."

The patent application was filed on May 3, 2012 (13/463,659).

Container marker scheme
for reducing write amplification in SSDs
IBM Corp., Armonk, NY, has been assigned a patent (8,463,983) developed by Evangelos S. Eleftheriou, Robert Haas, and Xiaoyu Hu, Ruschlikon, Switzerland, for a "container marker scheme for reducing write amplification in solid state devices."

The abstract of the patent published by the U.S. Patent and Trademark Office states: "A solid state storage device and method are provided. Multiple blocks are configured as storage memory for a solid state storage device, and each block includes multiple pages. A controller is configured to operate the solid state storage device. A free block of the multiple blocks is assigned a marker level by the controller. For a particular page of the multiple pages, each particular page of data is written to a block of the multiple blocks with a marker level corresponding to a level of dynamicity calculated by the controller for that particular page."

The patent application was filed on Sept. 15, 2009 (12/559,959).

Recovery of failed disks in RAID
IBM Corp., Armonk, NY, has been assigned a patent (8,464,090) developed by four co-inventors for a "recovery of failed disks in an array of disks."

The co-inventors are Abhinay R. Nagpal, Hadapsar, India, Sandeep R. Patil, Elmsford, NY, Sri Ramanathan, Lutz, FL, and Matthew B. Trevathan, Kennesaw, GA.

The abstract of the patent published by the U.S. Patent and Trademark Office states: "A disk recovery system and method is provided for recovering data from a redundant array of independent (or inexpensive) disks. The method is implemented in a computer infrastructure having computer executable code tangibly embodied on a computer readable storage medium having programming instructions. The computer readable storage medium is operable to: determine input/output (I/O) characteristics of one or more disks in an array; monitor the one or more disks in the array to determine when any of the one or more of the disks have failed in the array; and automatically rebuild the failed disk from a standby pool by nominating a disk in the standby pool based on the I/O characteristics of the one or more failed disks prior to failure."

The patent application was filed on Sept. 21, 2010 (12/887,007).

Replicating data
IBM Corp., Armonk, NY, has been assigned a patent (8,463,746) developed by six co-inventors for a "method and system for replicating data."

The co-inventors are Lisa J. Gundy, Beth A. Peterson, Alfred E. Sanchez, , David M. Shackelford, Tucson, AZ, Warren K. Stanley, Loveland, CO, and John G. Thompson, Tucson, AZ.

The abstract of the patent published by the U.S. Patent and Trademark Office states: "Methods, computing environments, and computer program products are provided for replicating data. Point-in-time copy parameters are generated at a primary host site. A side file entry having the point-in-time copy parameters and a track information field (TIF) associated with the point-in-time copy parameters is generated. The side file entry is transferred to a secondary host site. A point-in-time copy process is performed at the secondary host site based on the point-in-time copy parameters and the TIF."

The patent application was filed on March 30, 2011 (13/075,896).