Apparatus for maintaining duplicate copies of information stored in fault-tolerant computer main memories is disclosed. A non write-through cache memory associated with each of the system's processing elements stores computations generated by that processing element. At a context switch, the stored information is sequentially written to two separate main memory units. A separate status area in main memory is updated by the processing element both before and after each writing operation so that a fault occurring during data processing or during any storage operation leaves the system with sufficient information to be able to reconstruct the data without loss of integrity. To efficiently transfer information between the cache memory and the system main memories without consuming a large amount of processing time at context switches, a block status memory associated with the cache memory contains an entry for each data block in the cache memory. The entry indicates whether the corresponding data block has been modified during data processing or written with computational data from the processing element. The storage operations are carried out by high-speed hardware which stores only the modified data blocks. Additional special-purpose hardware simultaneously invalidates all cache memory entries so that a new task can be loaded and started.

Memory back-up system

A mechanism for maintaining a consistent, periodically updated state in main memory without constraining normal computer operation is provided, thereby enabling a computer system to recover from faults without loss of data or processing continuity. In this invention, a first computer includes a processor and input/output elements connected to a main memory subsystem including a primary element. A second computer has a remote checkpoint memory element, which may include one or more buffer memories and a shadow memory, which is connected to the main memory subsystem of the first computer. During normal processing, an image of data written to the primary memory element is captured by the remote checkpoint memory element. When a new checkpoint is desired (thereby establishing a consistent state in main memory to which all executing applications can safely return following a fault), the data previously captured is used to establish a new checkpointed state in the second computer. In case of failure of the first computer, the second computer can be restarted to operate from the last checkpoint established for the first computer. This structure and protocol can guarantee a consistent state in main memory, thus enabling fault-tolerant operation.

Remote checkpoint memory system and protocol for fault-tolerant computer system

Apparatus for maintaining duplicate copies of information stored in fault-tolerant computer main memories is disclosed A non write-through cache memory associated with each of the system's processing elements stores computations generated by that processing element At a context switch, the stored information is sequentially written to two separate main memory units A separate status area in main memory is updated by the processing element both before and after each writing operation so that a fault occurring during data processing or during any storage operation leaves the system with sufficient information to be able to reconstruct the data without loss of integrity To efficiently transfer information between the cache memory and the system main memories without consuming a large amount of processing time at context switches, a block status memory associated with the cache memory contains an entry for each data block in the cache memory The entry indicates whether the corresponding data block has been modified during data processing or written with computational data from the processing element The storage operations are carried out by high-speed hardware which stores only the modified data blocks Additional special-purpose hardware simultaneously invalidates all cache memory entries so that a new task can be loaded and started

Memory back up system with one cache memory and two physically separated main memories

A mechanism for maintaining a consistent state in main memory without constraining normal computer operation is provided, thereby enabling a computer system to recover from faults without loss of data or processing continuity. In a typical computer system, a processor and input/output elements are connected to a main memory via a memory bus. A shadow memory element, which includes a buffer memory and a main storage element, is also attached to this memory bus. During normal processing, data written to primary memory is also captured by the buffer memory of the shadow memory element. When a checkpoint is desired (thereby establishing a consistent state in main memory to which all executing applications can safely return following a fault), the data previously captured in the buffer memory is then copied to the main storage element of the shadow memory element. This structure and protocol can guarantee a consistent state in main memory, thus enabling fault-tolerant operation.

Main memory system and checkpointing protocol for fault-tolerant computer system

A method for insuring data consistency between a plurality of individual processor cache memories and the main memory in a multi-processor computer system is provided which is capable of (1) detecting when one of a set of predefined data inconsistency states occurs as a data transaction request is being processed, and (2) correcting the data inconsistency states so that the operation may be executed in a correct and consistent manner. In particular, the method is adapted to address two kinds of data inconsistency states: (1) A request for a operation from a system unit to main memory when the location to be written to is present in the cache of some processor unit-in such a case, data in the cache is "stale" and the data inconsistency is avoided by preventing the associated processor from using the "stale" data; and (2) when a read operation is requested of main memory by a system unit and the location to be read may be written or has already been written in the cache of some processor--in this case, the data in main memory is "stale" and the data inconsistency is avoided by insuring that the data returned to the requesting unit is the updated data in the cache. The presence of one of the above-described data inconsistency states is detected in a SCU-based multi-processing system by providing the SCU with means for maintaining a copy of the cache directories for each of the processor caches. The SCU continually compares address data accompanying memory access requests with what is stored in the SCU cache directories in order to determine the presence of predefined conditions indicative of data inconsistencies, and subsequently executes corresponding predefined fix-up sequences.

Scheme for insuring data consistency between a plurality of cache memories and the main memory in a multi-processor system

A fast cache flush mechanism includes, associated with the cache, an auxiliary portion (termed a flush count memory) that references a flush counter during the addressing of the cache. This flush counter preferably has a count capacity of the same size as the number of memory locations in the cache. Whenever the cache is updated, the current value of the flush counter is written into the location in the flush count memory associated with the memory location in the cache pointed to by an accessing address, and the valid bit is set. Whenever it is desired to flush the cache, the contents of the flush counter are changed (e.g. incremented) to a new value which is then written as the new cache index into the location of the flush count memory associated with that flush count, and the associated valid bit is cleared or reset. Any access to the cache by the address requires that the cache index in the associated flush count memory location match the current contents of the flush counter and that the valid bit be set. When the cache is flushed by the above procedure, these conditions cannot be fulfilled, since the current contents of the flush counter do not match any cache index or the valid bit has been reset. As a result, for each addressed memory location that has not been accessed since the last cache flush command (corresponding to the latest incrementing of the flush counter), the total contents of that memory location (i.e. data, cache index and validity bit) are updated in the manner described above. Through this procedure, once the contents of the flush counter have recycled back to a previous value, it is guaranteed that each memory location in the cache will have been flushed and, in many instances, updated with valid data.

Josephus Riffe

Papers

Fast cache flush mechanism