High performance storage unit
AssigneeSperry Corporation
InventorJames H. Scheuneman
Life of this patent
Estimating the term…
Loading patent…
AssigneeSperry Corporation
InventorJames H. Scheuneman
Estimating the term…
A large capacity (8 memory banks of 524K error-corrected 36 bit words stored) high performance (latency as low as 240 nanoseconds, 12.8 gigabits/second aggregate data transfer capability with up to 11.4 gigabits/second utilized) pipelined (8 deep request pipeline) random access memory store simultaneously (to the limit of bank addressing conflicts) services intermixed requests from an internal exerciser plus ported requestors (up to 10) of plural types (3 types), which requestors are not of the same interface cycle time (30 nsec vs. 60 nsec). Furthermore, to such nonuniform interface cycle times, the bit-width of the data transfer interfaces (ports) to the requestors of plural types is also not uniform, but is actually wider (4 interface words of 36 bits each=144 bits) to faster (30 nanosecond) requestors than is that data transfer bit-width (2 interface words=72 bits) to slower (60 nsec) requestors. Three differing individual requestor bandwidths=data-transfer-bit width/data-transfer-period (144 bits/30 nanosecond, 72 bits/120 nanosecond or 248 bits/240 nanosecond) are intermixedly and simultaneously (to the limit of addressing possibility) supported on 8 memory banks each of which does output 144 bits per 90 nanoseconds.
A memory resource synchronously communicative with a plurality of requesters for the reading and writing of addressable data stores operative in a manner such that said memory resource does synchronously communicate with at least one of said plurality of requestors at a first interface communication cycle time T.sub.x while simultaneously synchronously communicating with at least another one of said plurality of requesters at a different second interface communication cycle time T.sub.y, T.sub.x =T.sub.y, said interface communication cycle time being defined as that time between successive synchronous receipts of a request signal plus accompanying address and data and control information from a single requestor, which is equivalent to defining said interface communication cycle time as that time interval during which synchronous communication of information transpires, said memory resource operative for said synchronously communicating at a plurality of different interface communication cycle times comprising:
A memory resource synchronously communicative with each such plurality of requestors for the reading and writing of addressable data stores operative in a manner such that said memory resource does always and invariantly synchronously communicate for the reading and writing of common addressable data stores with at least one of said plurality of requestors at a first interface bit-width W.sub.x while simultaneously synchronously communicating with at least another one of said plurality of requestors at a second interface bit-width W.sub.y, W.sub.x =W.sub.y, said interface bit-width being defined as the number of bits of data which are communicated upon each synchronous interface communication cycle time to each of said plurality of requestors and said memory resource does perform reading and writing in common on a priority basis to all said plurality of requestors collectively, meaning that both said at least one of said plurality of requestors and said another one of said plurality of requestors can read and write, on a priority basis, the same locations of said common addressable data stores, said memory resource operative for said synchronously communicating on a priority basis at a plurality of interface communication cycle times comprising:
Dependencies are inferred from the claim text.