Abstract: Methods, systems and computer program products for providing access to a spare memory array element (“MAE”) are provided. Aspects include storing a row number a column number associated with a defective MAE of a plurality of MAEs. The plurality of MAEs are logically arranged in a plurality of rows and a plurality of columns. Aspects also include receiving a command to access a cache line. The cache line corresponds to a selected row of MAEs of the plurality of MAEs. Responsive to determining that the selected row matches the row number that is associated with the defective MAE, aspects include activating one or more column shifters to prevent access to the defective MAE and provide access to a spare MAE when accessing the cache line. The activation of the one of more column shifters is based on the column number that is associated with the defective MAE.

Abstract: Methods, systems and computer program products for evacuating memory from a drawer in a live multi-node system are provided. Aspects include placing a first drawer into an evacuation mode. The evacuation mode includes a cessation of non-evacuation operations and provides for a transfer of data stored by memory of the first drawer to a destination drawer using dynamic memory reallocation (DMR). Aspects also include transmitting a store request by the first drawer to the destination drawer. The store request represents a request to transfer the data stored by the memory of the first drawer to the destination drawer for storage by the destination drawer. Aspects also include transmitting the data stored by the memory of the first drawer to the destination drawer. The data is transmitted by the first drawer using a local pool of fetch/store controllers.

Abstract: Embodiments of the present invention are directed to a computer-implemented method for cache eviction. The method includes detecting a first data in a shared cache and a first cache in response to a request by a first processor. The first data is determined to have a mid-level cache eviction priority. A request is detected from a second processor for a same first data as requested by the first processor. However, in this instance, the second processor has indicated that the same first data has a low-level cache eviction priority. The first data is duplicated and loaded to a second cache, however, the data has a low-level cache eviction priority at the second cache.

Abstract: Methods and systems for cache management are provided. Aspects include providing a drawer including a plurality of clusters, each of the plurality of clusters including a plurality of processor each having one or more cores, wherein each of the one or more cores shares a first cache memory, providing a second cache memory shared among the plurality of clusters, and receiving a cache line request from one of the one or more cores to the first cache memory, wherein the first cache memory sends a request to a memory controller to retrieve the cache line from a memory, store the cache line in the first cache memory, create a directory state associated with the cache line, and provide the directory state to the second cache memory to create a directory entry for the cache line.

Abstract: Processing simultaneous data requests regardless of active request in the same addressable index of a cache. In response to the cache miss in the given congruence, if the number of other compartments in the given congruence class that have an active operation is less than a predetermined threshold, setting a Do Not Cast Out (DNCO) pending indication for each of the compartments that have an active operation in order to block access to each of the other compartments that have active operations and, if the number of other compartments in the given congruence class that have an active operation is not less than a predetermined threshold, blocking another cache miss from occurring in the compartments of the given congruence class by setting a congruence class block pending indication for the given congruence class in order to block access to each of the other compartments of the given congruence class.

Abstract: Processing simultaneous data requests regardless of active request in the same addressable index of a cache. In response to the cache miss in the given congruence, if the number of other compartments in the given congruence class that have an active operation is less than a predetermined threshold, setting a Do Not Cast Out (DNCO) pending indication for each of the compartments that have an active operation in order to block access to each of the other compartments that have active operations and, if the number of other compartments in the given congruence class that have an active operation is not less than a predetermined threshold, blocking another cache miss from occurring in the compartments of the given congruence class by setting a congruence class block pending indication for the given congruence class in order to block access to each of the other compartments of the given congruence class.

Abstract: Provided are systems, methods, and media for simultaneous, non-atomic request processing of snooped operations of a broadcast scope within a SMP system. An example method includes detecting, by a first controller, based on a set of coherency resolution conditions, whether there are coherency resolution problems between two snooped operations. The method includes in response to detecting, by the first controller, that coherency resolution problems will not result, transmitting, from the first controller to a second controller, an indication signal indicating that coherency resolution problems will not result from the operation. The set of coherency resolution conditions includes: (a) detecting that a second operation of the two snooped operations operation is of a predetermined type, (b) detecting at time of snooping of the second operation that a directory state does not allow for exclusive data, and (c) detecting that the first controller has started committing to an update.

Abstract: Methods and systems for cache management are provided. Aspects include providing a drawer including a plurality of clusters, each of the plurality of clusters including a plurality of processor each having one or more cores, wherein each of the one or more cores shares a first cache memory, providing a second cache memory shared among the plurality of clusters, and receiving a cache line request from one of the one or more cores to the first cache memory, wherein the first cache memory sends a request to a memory controller to retrieve the cache line from a memory, store the cache line in the first cache memory, create a directory state associated with the cache line, and provide the directory state to the second cache memory to create a directory entry for the cache line.

Abstract: Embodiments of the present invention are directed to a computer-implemented method for cache eviction. The method includes detecting a first data in a shared cache and a first cache in response to a request by a first processor. The first data is determined to have a mid-level cache eviction priority. A request is detected from a second processor for a same first data as requested by the first processor. However, in this instance, the second processor has indicated that the same first data has a low-level cache eviction priority. The first data is duplicated and loaded to a second cache, however, the data has a low-level cache eviction priority at the second cache.

Abstract: Methods, systems and computer program products for providing access to a spare memory array element (“MAE”) are provided. Aspects include storing a row number a column number associated with a defective MAE of a plurality of MAEs. The plurality of MAEs are logically arranged in a plurality of rows and a plurality of columns. Aspects also include receiving a command to access a cache line. The cache line corresponds to a selected row of MAEs of the plurality of MAEs. Responsive to determining that the selected row matches the row number that is associated with the defective MAE, aspects include activating one or more column shifters to prevent access to the defective MAE and provide access to a spare MAE when accessing the cache line. The activation of the one of more column shifters is based on the column number that is associated with the defective MAE.

Abstract: Provided are systems, methods, and media for simultaneous, non-atomic request processing of snooped operations of a broadcast scope within a SMP system. An example method includes detecting, by a first controller, based on a set of coherency resolution conditions, whether there are coherency resolution problems between two snooped operations. The method includes in response to detecting, by the first controller, that coherency resolution problems will not result, transmitting, from the first controller to a second controller, an indication signal indicating that coherency resolution problems will not result from the operation. The set of coherency resolution conditions includes: (a) detecting that a second operation of the two snooped operations operation is of a predetermined type, (b) detecting at time of snooping of the second operation that a directory state does not allow for exclusive data, and (c) detecting that the first controller has started committing to an update.

Abstract: Methods, systems and computer program products for evacuating memory from a drawer in a live multi-node system are provided. Aspects include placing a first drawer into an evacuation mode. The evacuation mode includes a cessation of non-evacuation operations and provides for a transfer of data stored by memory of the first drawer to a destination drawer using dynamic memory reallocation (DMR). Aspects also include transmitting a store request by the first drawer to the destination drawer. The store request represents a request to transfer the data stored by the memory of the first drawer to the destination drawer for storage by the destination drawer. Aspects also include transmitting the data stored by the memory of the first drawer to the destination drawer. The data is transmitted by the first drawer using a local pool of fetch/store controllers.

Abstract: A technique for cache coherency is provided. A cache controller selects a first set from multiple sets in a congruence class based on a cache miss for a first transaction, and places a lock on the entire congruence class in which the lock prevents other transactions from accessing the congruence class. The cache controller designates in a cache directory the first set with a marked bit indicating that the first transaction is working on the first set, and the marked bit for the first set prevents the other transactions from accessing the first set within the congruence class. The cache controller removes the lock on the congruence class based on the marked bit being designated for the first set, and resets the marked bit for the first set to an unmarked bit based on the first transaction completing work on the first set in the congruence class.

Abstract: A technique for cache coherency is provided. A cache controller selects a first set from multiple sets in a congruence class based on a cache miss for a first transaction, and places a lock on the entire congruence class in which the lock prevents other transactions from accessing the congruence class. The cache controller designates in a cache directory the first set with a marked bit indicating that the first transaction is working on the first set, and the marked bit for the first set prevents the other transactions from accessing the first set within the congruence class. The cache controller removes the lock on the congruence class based on the marked bit being designated for the first set, and resets the marked bit for the first set to an unmarked bit based on the first transaction completing work on the first set in the congruence class.