International Business Machines Corporation

Globalfoundries Inc.

Gordon J Cockburn

Adrian John Hawes

Ian David Judd

John P Wilkinson

Carlos Francisco Fuente

Patrick Allen Buckland

Christopher B E Beeken

Joanna Katharine Brown

Christopher Canto

Vinay V Shah

Bernard John Grainger

William A Short

David James Medhurst

Roger G Cath

A first data storage device receives a first write request from a host device to update a first subset of the data regions and sends the first write request to a second data storage device. Prior to receiving a notification from the second device that it has been updated in accordance with the first write request, the first device receives a second write request from the second device to update a second subset of the data regions and one of the two devices is selected. If the first device is selected, the first device updates only its data regions that are in the second subset but not in the first subset according to the second write request. In the case that the second device is selected, the first device updates its data regions that are in the second subset according to the second write request.

Replicating data in a data storage system

Disclosed is a system for replicating data. The system may comprise a plurality of nodes preferably organized in groups with one of the nodes acting as a coordinator node. The nodes are configured to receive write requests from an external server and to apply these write requests to a data storage source of the data storage system. The write requests typically belong to a batch of independent write actions identified by a batch sequence number. Each node stores the write request in non-volatile memory with the coordinator node monitoring which batches are secured in their entirety in non-volatile memory. The coordinator node authorizes all other nodes to sequentially replicate the write requests in their non-volatile memory to the data storage source for all writes up to the highest batch sequence number for which all writes have been secured in non-volatile memory.

Robust data replication

Provided are a method, system, and article of manufacture for synchronizing controller firmware download. A master device controller receives a code download and sends the code download to a subordinate device controller. The master device controller requests a subordinate code status from the subordinate device controller indicating a status of the code download at the subordinate device controller and stores the received code download and a master code status for the download. The master device controller receives the subordinate code status from the subordinate device controller and compares the subordinate code status with the master code status to determine whether both are valid and at same code level. The master device controller updates the master code image with the code download and coordinating the update of the code download to the subordinate code image in response to determining that the master and subordinate code status are both valid and at the same code level.

Synchronizing controller firmware download

Apparatus for parsing and tokenizing a data stream comprises: a storage component to store a history buffer containing an unencoded version of a previously encoded string; a comparison component to compare a string from the input data stream with the unencoded version of at least one previously encoded string; a second storage component store: an indicator that at least two matches were found by the first comparison component, and tokens corresponding to the matches; a summing component to sum potential token lengths to provide total potential token lengths; a second comparison component to compare total potential token lengths; a selection component to select a match corresponding to a shortest total token length to represent the string from said input data stream; and an emitting component for emitting tokens representing the match corresponding to the shortest total token length. The tokens may be used in, for example, compression or encryption.

Data parsing and tokenizing apparatus, method and program

A method () and arrangement for LZ1 compression of a data string where multiple input symbols are compared in parallel with the history buffer by: holding in an input buffer () a first sequence of bytes of the data string; holding in a history buffer () a second sequence of bytes of the data string; comparing (), in matrix comparison means coupled to the input buffer and the history buffer and having a plurality of rows and columns of comparison units (), bytes held in the input buffer with bytes held in the history buffer, bytes of the history buffer being coupled to diagonally displaced comparison units in the matrix comparison means; detecting () in each of the rows the column in which a largest number of consecutive byte matches has occurred at the comparison unit in that row and preceding comparison units in the same column; and encoding () as a token a sequence of matched bytes detected in the step of detecting ().

Method and arrangement for data compression according to the LZ77 algorithm

An apparatus for performing data compression is disclosed. A circuit () within a comparison unit () of a comparison matrix () performs LZ1 compression of a data string by comparing bytes held in an input buffer () with bytes held in a history buffer (). A group of logic gate stages () is connected in series with each other. Each of the logic gate stages produces a carry value that is passed to one of the output of the comparison unit and another logic gate stages. The product of the number stages in the logic gate stages and the number logic gates in each of the logic gate stages is less than the number of logic gates required for an equivalent circuit having a single logic circuit stage.

Circuit and method for use in data compression

A method and arrangement are described for decompressing three or more bytes per processor cycle from a stream of compressed data using a processing pipeline, in which the compressed data is represented by tokens of varying and unknown length by accepting as input the stream of token data, partially decoding a token from the token data to determine a boundary position of the token; and priming the processing pipeline with the token and a length marker indicating the boundary position. Literal data is detected and output directly, copy pointers are checked to find their targets and resolved either from the history buffer or from the in-process contents of the pipeline.

Multi-byte Lempel-Ziv 1(LZ1) decompression

A data storage subsystem and method for transferring data from a storage subsystem to a connected host data processing system are described. The subsystem comprises a device controller connected to one or more direct access storage devices e.g. disk drives. The host data processing system issues data transfer commands to the subsystem to initiate transfer of data between the host processing system and the device(s) associated with the data storage subsystem. Read/write data is transferred directly from device to host via a buffer controller. For a read operation, the read command from the host data processing system specifies the data to be transferred and the start address in host memory to which the data should be sent. The device controller of the data storage subsystem is capable of respecifying or amending the start address specified by the host in the rad command. This provides a performance bonefit for split data transfers. In addition, if an error occurs during a read operation, the device controller of the data storage subsystem can specify the host address to which the replacement data should be sent.

Growing community of inventors

Southampton, United Kingdom

Gordon J Cockburn