Abstract: A system for providing a cluster-wide system clock in a multi-tiered full graph (MTFG) interconnect architecture are provided. Heartbeat signals transmitted by each of the processor chips in the computing cluster are synchronized. Internal system clock signals are generated in each of the processor chips based on the synchronized heartbeat signals. As a result, the internal system clock signals of each of the processor chips are synchronized since the heartbeat signals, that are the basis for the internal system clock signals, are synchronized. Mechanisms are provided for performing such synchronization using direct couplings of processor chips within the same processor book, different processor books in the same supernode, and different processor books in different supernodes of the MTFG interconnect architecture.

Abstract: A multiprocessor computer system in which snoop operations of the caches are synchronized to allow the issuance of a cache operation during a cycle which is selected based on the particular manner in which the caches have been synchronized. Each cache controller is aware of when these synchronized snoop tenures occur, and can target these cycles for certain types of requests that are sensitive to snooper retries, such as kill-type operations. The synchronization may set up a priority scheme for systems with multiple interconnect buses, or may synchronize the refresh cycles of the DRAM memory of the snooper's directory. In another aspect of the invention, windows are created during which a directory will not receive write operations (i.e., the directory is reserved for only read-type operations). The invention may be implemented in a cache hierarchy which provides memory arranged in banks, the banks being similarly synchronized.

Abstract: In response to a need to initiate a global operation, a bus master within a multiprocessor system issues a combined token and operation request on a bus coupled to the bus master. The combined token and operation request solicits a token required to complete the global operation and identifies the global operation to be processed with the token, if granted. Upon receiving a combined response acknowledging both the token and operation portions of the combined request, the bus master treats the global operation as complete. If a combined response acknowledging the token portion of the combined request but retrying the operation portion (i.e., at least one snooper is busy processing a previous global operation), the bus master issues an operation request (only) for the operation portion of the combined request.

Abstract: A multiprocessor computer system in which snoop operations of the caches are synchronized to allow the issuance of a cache operation during a cycle which is selected based on the particular manner in which the caches have been synchronized. Each cache controller is aware of when these synchronized snoop tenures occur, and can target these cycles for certain types of requests that are sensitive to snooper retries, such as kill-type operations. The synchronization may set up a priority scheme for systems with multiple interconnect buses, or may synchronize the refresh cycles of the DRAM memory of the snooper's directory. In another aspect of the invention, windows are created during which a directory will not receive write operations (i.e., the directory is reserved for only read-type operations). The invention may be implemented in a cache hierarchy which provides memory arranged in banks, the banks being similarly synchronized.

Abstract: A multiprocessor computer system in which snoop operations of the caches are synchronized to allow the issuance of a cache operation during a cycle which is selected based on the particular manner in which the caches have been synchronized. Each cache controller is aware of when these synchronized snoop tenures occur, and can target these cycles for certain types of requests that are sensitive to snooper retries, such as kill-type operations. The synchronization may set up a priority scheme for systems with multiple interconnect buses, or may synchronize the refresh cycles of the DRAM memory of the snooper's directory. In another aspect of the invention, windows are created during which a directory will not receive write operations (i.e., the directory is reserved for only read-type operations). The invention may be implemented in a cache hierarchy which provides memory arranged in banks, the banks being similarly synchronized.

Abstract: A multiprocessor computer system in which snoop operations of the caches are synchronized to allow the issuance of a cache operation during a cycle which is selected based on the particular manner in which the caches have been synchronized. Each cache controller is aware of when these synchronized snoop tenures occur, and can target these cycles for certain types of requests that are sensitive to snooper retries, such as kill-type operations. The synchronization may set up a priority scheme for systems with multiple interconnect buses, or may synchronize the refresh cycles of the DRAM memory of the snooper's directory. In another aspect of the invention, windows are created during which a directory will not receive write operations (i.e., the directory is reserved for only read-type operations). The invention may be implemented in a cache hierarchy which provides memory arranged in banks, the banks being similarly synchronized.

Abstract: In response to a need to initiate one or more global operations, a bus master within a multiprocessor system issues a combined token and operation request in a single bus transaction on a bus coupled to the bus master. The combined token and operation request solicits a single existing token required to complete the global operations within the multiprocessor system and identifies the first of the global operations to be processed with the token, if granted. Once a bus master is granted the token, no other bus master will be granted the token until the current token owner explicitly requests release. The current token owner repeats the combined token and operation request for each global operation which needs to be initiated and, on the last global operation, issues a combined request with an explicit release. Acknowledgement of the combined request with release implies release of the token for use by other bus masters.

Abstract: In response to a need to initiate a global operation, a bus master within a multiprocessor system issues a combined token and operation request on a bus coupled to the bus master. The combined token and operation request solicits one of a plurality of tokens required to complete the global operation and identifies the global operation to be processed with the token, if granted. Bus snoopers contain a number of snooper queues for global operations equal to the number of global operation tokens employed within the multiprocessor system. A bus snooper, upon detecting a combined token and operation request, begins speculatively processing the operation if the snooper is not already busy. Before completing the operation, the snooper watches for a combined response with a token number acknowledging either the combined request or a subsequent token request from the same processor, which indicates that the originating bus master has been granted a token for completing a global operation.

Abstract: In cancelling the cast out portion of a combined operation including a data access related to the cast out, the combined response logic explicitly directs a horizontal storage device at the same level as the storage device initiating the combined operation to allocate and store either the cast out or target data. A horizontal storage device having available space—i.e., an invalid or modified data element in a congruence class for the victim—stores either the target or the cast out data for subsequent access by an intervention. Cancellation of the cast out thus defers any latency associated with writing the cast out victim to system memory while maximizing utilization of available storage with acceptable tradeoffs in data access latency.

Abstract: Serialization of global operations within a multiprocessor system is achieved utilizing a single token, requiring a bus master to acquire the token for completion of each individual global operation initiated by that bus master. A combined token and operation request, in which a token request and an operation request are transmitted in a single bus transaction, is employed once for a global operation, to initiate the global operation for the first time. A token manager determines whether the token is available or checked out and responds to the token portion of the combined request. Snoopers respond to the operation portion of the combined request depending on whether they are busy. If the entire combined request is retried, a token request (only) is employed to request the token and, when the token is acquired, an operation request (only) is employed to request the operation.

Abstract: Following a cache miss by an operation, the address for the operation is transmitted on the bus coupling the cache to lower levels of the storage hierarchy. A portion of the address including the index field is transmitted during a first bus cycle, and may be employed to begin directory lookups in lower level storage devices before the address tag is received. The remainder of the address is transmitted during subsequent bus cycles, which should be in time for address tag comparisons with the congruence class elements. To allow multiple directory lookups to be occurring concurrently in a pipelined directory, a portion of multiple addresses for several data access operations, each portion including the index field for the respective address, may be transmitted during the first bus cycle or staged in consecutive bus cycles, with the remainders of each address—including the cache tags—transmitted during the subsequent bus cycles.

Abstract: Only a single snooper queue for global operations within a multiprocessor system is implemented within each bus snooper, controlled by a single token allowing completion of one operation. A bus snooper, upon detecting a combined token and operation request, begins speculatively processing the operation if the snooper is not already busy. The snooper then watches for a combined response acknowledging the combined request or a subsequent token request from the same processor, which indicates that the originating processor has been granted the sole token for completing global operations, before completing the operation. When processing an operation from a combined request and detecting an operation request (only) from a different processor, which indicates that another processor has been granted the token, the snooper suspends processing of the current operation and begins processing the new operation.

Abstract: Serialization of global operations within a multi-processor system is achieved utilizing a plurality of tokens each permitting completion of a single global operation, requiring a bus master to acquire the token for completion of each individual global operation initiated by that bus master. A combined token and operation request, in which a token request and an operation request are transmitted in a single bus transaction, is employed once for a global operation, to initiate the global operation for the first time. A token manager determines whether a token is available or all are checked out and responds to the token portion of the combined request. Snoopers respond to the operation portion of the combined request depending on whether they are busy. If at least the token portion of the combined request is acknowledged or if a token request is acknowledged, the combined response will include a token number for the token granted to the bus master initiating the global operation.

Abstract: A multiprocessor computer system in which snoop operations of the caches are synchronized to allow the issuance of a cache operation during a cycle which is selected based on the particular manner in which the caches have been synchronized. Each cache controller is aware of when these synchronized snoop tenures occur, and can target these cycles for certain types of requests that are sensitive to snooper retries, such as kill-type operations. The synchronization may set up a priority scheme for systems with multiple interconnect buses, or may synchronize the refresh cycles of the DRAM memory of the snooper's directory. In another aspect of the invention, windows are created during which a directory will not receive write operations (i.e., the directory is reserved for only read-type operations). The invention may be implemented in a cache hierarchy which provides memory arranged in banks, the banks being similarly synchronized.

Abstract: A multiprocessor computer system in which snoop operations of the caches are synchronized to allow the issuance of a cache operation during a cycle which is selected based on the particular manner in which the caches have been synchronized. Each cache controller is aware of when these synchronized snoop tenures occur, and can target these cycles for certain types of requests that are sensitive to snooper retries, such as kill-type operations. The synchronization may set up a priority scheme for systems with multiple interconnect buses, or may synchronize the refresh cycles of the DRAM memory of the snooper's directory. In another aspect of the invention, windows are created during which a directory will not receive write operations (i.e., the directory is reserved for only read-type operations). The invention may be implemented in a cache hierarchy which provides memory arranged in banks, the banks being similarly synchronized.

Abstract: Serialization of global operations within a multiprocessor system is achieved utilizing a single token, requiring a bus master to acquire the token for completion of one or more global operations to be initiated by that bus master. A combined token and operation request, in which a token request and an operation request are transmitted in a single bus transaction, is employed once for a global operation, to initiate the global operation for the first time. A token manager determines whether the token is available and released and, if available but not released, whether the token is checked out to the bus master originating the combined token and operation request. If the token is available and released or is available and was last checked out to the bus master originating the combined token and operation request, the token manager acknowledges to the token portion of the combined request; otherwise the token manager retries the token portion of the combined request.

Abstract: A multiprocessor computer system in which snoop operations of the caches are synchronized to allow the issuance of a cache operation during a cycle which is selected based on the particular manner in which the caches have been synchronized. Each cache controller is aware of when these synchronized snoop tenures occur, and can target these cycles for certain types of requests that are sensitive to snooper retries, such as kill-type operations. The synchronization may set up a priority scheme for systems with multiple interconnect buses, or may synchronize the refresh cycles of the DRAM memory of the snooper's directory. In another aspect of the invention, windows are created during which a directory will not receive write operations (i.e., the directory is reserved for only read-type operations). The invention may be implemented in a cache hierarchy which provides memory arranged in banks, the banks being similarly synchronized.

Abstract: Upon snooping a combined data access and castout/deallocate operation initiating by a horizontal storage device, snoop logic determines, from coherency state information appended to either the combined operation or the combined response to the operation, whether the coherency state of the victim may be upgraded within the subject storage device. If so, the coherency state is upgraded to improve global data storage management. For instance, a cache line may be upgraded from the shared coherency state to the exclusive coherency state to improve data storage management under a given replacement policy.

Abstract: An effectively “conditional”, cast out operation or cast out portion of a combined operation including a related data access may be cancelled by the combined response to the operation. The combined response logic receives coherency state and/or LRU position information for cache lines corresponding to the cast out victim within snoopers and vertically in-line storage. The combined response logic may also receive information regarding the presence of shared or invalid cache lines in snoopers or lower level storage within the congruence class for the victim, or information regarding the read-once nature of the data access target. Based on these responses, the combined response logic determines whether the cast out should be cancelled and, if so, selects and drives the appropriate combined response code.

Abstract: Upon snooping a combined data access and cast out/deallocate operation initiating by a horizontal storage device, snoop logic determines appropriate responses to both the data access and the cast out/deallocate based upon the presence and coherency state of the target of the data access within a corresponding storage device, the presence and coherency state of the victim of the cast out/deallocate within the corresponding storage device, and the presence of an invalid entry within the corresponding storage device in a congruence class including both the target and the victim. The appropriate responses are “merged”, transmitted together in response to the combined operation as either a single response code or discrete response codes within a single response. The coherency state and LRU position of the selected victim for the cast out/deallocate portion of the combined operation may also be appended to the response to facilitate data storage management.