摘要:
A system for managing distributed storage includes a collection of one or more compute sites that are connected and that perform a joint computation. A witness node is connected to the one or more compute sites. A membership service allows a change in collection membership, in response to a failure, to continue the joint computation without causing data inconsistency or corruption. If the membership service is unable to reach a decision on the change in collection membership in response to the failure, the witness node controls the decision on the change in collection membership to continue the joint computation. A distributed storage system provided by the system described herein may advantageously meet consistency guarantees and maximize data access even when individual compute sites and/or their interconnects fail.
摘要:
Handling asynchronous data provided by at least two sources includes synchronizing data between the at least two sources, coordinating cycle switching between the at least two sources, where an asynchronous cycle of data from one of the sources corresponds to an asynchronous cycle of data from an other one of the sources, receiving asynchronous data corresponding to a data cycle from at least one of the sources, and saving at least some of the received data. A first portion of data for a particular cycle may be transmitted from one of the sources and a second portion of data for the particular cycle, different from the first portion of data, may be transmitted from an other one of the sources and the first portion and the second portion together may correspond to all of the data for the particular cycle.
摘要:
Setting a plurality of table entries in a storage device includes subdividing the table entries into a N tasks, placing each of the N tasks in a memory location disposed within the storage device and accessible by a plurality of internal devices, the plurality of the internal devices accessing the memory location to retrieve at least one of the N tasks, and each of the plurality of the internal devices setting table entries corresponding to at least one of the N tasks retrieved from the memory location. Setting table entries may also include setting logical device table entries to indicate corresponding tracks contain invalid data in connection with operation of remote data transfer between multiple storage devices. At least some of the internal devices may include devices for handling remote data transfer between multiple storage devices. At least some of the internal devices may be disk adapters and host adapters of the storage device. The memory location may correspond to a queue.
摘要:
Handling ordered writes to a data storage device includes receiving a first chunk of data into slots of a cache of the data storage device and using a hash table to provide a correlation between cache slots containing data from the first chunk of data and locations of a logical storage device of the data storage device. Handling ordered writes may also include maintaining a circularly linked list of pointers to cache slots containing the first chunk of data. Handling ordered writes may also include receiving a second chunk of data into slots of the cache of the data storage device and receiving a message indicating that all data corresponding to the first chunk of data has been transferred to the data storage device. Handling ordered writes may also include, in response to receiving the message, storing data from the first chunk on to the logical storage device. Handling ordered writes may also include using a hash table to provide a correlation between cache slots containing data from the second chunk of data and locations of a logical storage device of the data storage device.
摘要:
Ordering data writes includes at least some of a group of primary storage devices receiving a first plurality of data writes, causing a cycle switch for the group of primary storage devices where the first plurality of data writes are associated with a particular cycle on each primary storage device in the group, and at least some of the group of primary storage devices receiving a second plurality of writes after initiating the cycle switch where all of the second plurality of writes are associated with a cycle different from the particular cycle on each primary storage device. Writes to the group begun after initiating the cycle switch may not complete until after the cycle switch has completed. Ordering data writes may also include, after completion of the cycle switch, each of the primary storage devices of the group initiating transfer of the first plurality of writes to a corresponding secondary storage device. Ordering data writes may also include, following each of the primary storage devices of the group completing transfer of the first plurality of writes to a corresponding secondary storage device, each of the primary storage devices sending a message to the corresponding secondary storage device.
摘要:
Transitioning from a first data transfer mode to a second data transfer mode that is different from the first data transfer mode includes a primary storage device receiving a plurality of data writes while being in the first data transfer mode, the primary storage device associating data writes begun after a first time and before a second time with a first chunk of data, the primary storage device associating data writes begun after the second time with a second chunk of data different from the first chunk of data, and the primary storage device transitioning to the second data transfer mode after the second time by first transferring writes associated with the first chunk of data to a secondary storage device after completion of all writes associated with the first chunk of data and then, after all writes associated with the first chunk of data have been transferred to the secondary storage device, transferring writes associated with the second chunk of data to the secondary storage device using the first data transfer mode, where writes begun after initiating transitioning are provided to the secondary storage device using the second data transfer mode. Transitioning from a first data transfer mode to a second data transfer mode that is different from the first data transfer mode may also include, following the primary storage device transferring all writes associated with the first chunk of data to the secondary storage device, the primary storage device sending a message to the secondary storage device, and, in response to receiving the message from the primary storage device, the secondary storage device storing the data writes associated with the first chunk of data.
摘要:
Ordering data writes include a host computer providing a plurality of data writes to a primary storage device, the primary storage device assigning a first sequence number to data writes begun after a first time and before a second time, the primary storage device assigning a second sequence number, different from the first sequence number, to data writes begun after the second time, in response to completion of all writes assigned the first sequence number, the primary storage device transferring writes assigned the first sequence number to a secondary storage device, and, in response to the primary storage device transferring all writes assigned the first sequence number to the secondary storage device, the primary storage device sending a commit message to the secondary storage device. Ordering data writes may also include, in response to receiving a commit message from the primary storage device, the secondary storage device storing the data writes having the first sequence number.
摘要:
Handling ordered writes to a data storage device includes receiving a first chunk of data into slots of a cache of the data storage device and using a hash table to provide a correlation between cache slots containing data from the first chunk of data and locations of a logical storage device of the data storage device. Handling ordered writes may also include maintaining a circularly linked list of pointers to cache slots containing the first chunk of data. Handling ordered writes may also include receiving a second chunk of data into slots of the cache of the data storage device and receiving a message indicating that all data corresponding to the first chunk of data has been transferred to the data storage device. Handling ordered writes may also include, in response to receiving the message, storing data from the first chunk on to the logical storage device. Handling ordered writes may also include using a hash table to provide a correlation between cache slots containing data from the second chunk of data and locations of a logical storage device of the data storage device.
摘要:
Ordering data writes include a host computer providing a plurality of data writes to a primary storage device, the primary storage device assigning a first sequence number to data writes begun after a first time and before a second time, the primary storage device assigning a second sequence number, different from the first sequence number, to data writes begun after the second time, in response to completion of all writes assigned the first sequence number, the primary storage device transferring writes assigned the first sequence number to a secondary storage device, and, in response to the primary storage device transferring all writes assigned the first sequence number to the secondary storage device, the primary storage device sending a commit message to the secondary storage device. Ordering data writes may also include, in response to receiving a commit message from the primary storage device, the secondary storage device storing the data writes having the first sequence number.
摘要:
Ordering data writes include a host computer providing a plurality of data writes to a primary storage device, the primary storage device assigning a first sequence number to data writes begun after a first time and before a second time, the primary storage device assigning a second sequence number, different from the first sequence number, to data writes begun after the second time, in response to completion of all writes assigned the first sequence number, the primary storage device transferring writes assigned the first sequence number to a secondary storage device, and, in response to the primary storage device transferring all writes assigned the first sequence number to the secondary storage device, the primary storage device sending a commit message to the secondary storage device. Ordering data writes may also include, in response to receiving a commit message from the primary storage device, the secondary storage device storing the data writes having the first sequence number.