Abstract: Systems, apparatuses, and methods for dynamic buffer management in multi-client token flow control routers are disclosed. A system includes at least one or more processing units, a memory, and a communication fabric with a plurality of routers coupled to the processing unit(s) and the memory. A router servicing multiple active clients allocates a first number of tokens to each active client. The first number of tokens is less than a second number of tokens needed to saturate the bandwidth of each client to the router. The router also allocates a third number of tokens to a free pool, with tokens from the free pool being dynamically allocated to different clients. The third number of tokens is equal to the difference between the second number of tokens and the first number of tokens. An advantage of this approach is reducing the amount of buffer space needed at the router.

Abstract: A computing system uses a memory for storing data, one or more clients for generating network traffic and a communication fabric with network switches. The network switches include centralized storage structures, rather than separate input and output storage structures. The network switches store particular metadata corresponding to received packets in a single, centralized collapsing queue where the age of the packets corresponds to a queue entry position. The payload data of the packets are stored in a separate memory, so the relatively large amount of data is not shifted during the lifetime of the packet in the network switch. The network switches select sparse queue entries in the collapsible queue, deallocate the selected queue entries, and shift remaining allocated queue entries toward a first end of the queue with a delay proportional to the radix of the network switches.

Abstract: A system for implementing load balancing schemes includes one or more processing units, a memory, and a communication fabric with a plurality of switches coupled to the processing unit(s) and the memory. A switch of the fabric determines a first number of streams on a first input port that are targeting a first output port. The switch also determines a second number of requestors, from all input ports, that are targeting the first output port. Then, the switch calculates a throttle factor for the first input port by dividing the first number of streams by the second number of streams. The switch applies the throttle factor to regulate bandwidth on the first input port for requestors targeting the first output port. The switch also calculates throttle factors for the other ports and applies the throttle factors when regulating bandwidth on the other ports.

Abstract: A computing system uses a memory for storing data, one or more clients for generating network traffic and a communication fabric with network switches. The network switches include centralized storage structures, rather than separate input and output storage structures. The network switches store particular metadata corresponding to received packets in a single, centralized collapsing queue where the age of the packets corresponds to a queue entry position. The payload data of the packets are stored in a separate memory, so the relatively large amount of data is not shifted during the lifetime of the packet in the network switch. The network switches select sparse queue entries in the collapsible queue, deallocate the selected queue entries, and shift remaining allocated queue entries toward a first end of the queue with a delay proportional to the radix of the network switches.

Abstract: Systems, apparatuses, and methods for using a scoreboard to track updates to configuration state registers are disclosed. A system includes one or more processing nodes, one or more memory devices, a plurality of configuration state registers, and a communication fabric coupled to the processing unit(s) and memory device(s). The system uses a scoreboard to track updates to the configuration state registers during run-time. Prior to a node going into a power-gated state, the system stores only those configuration state registers that have changed. This reduces the amount of data written to memory on each transition into power-gated state, and increases the amount of time the node can spend in the power-gated state. Also, configuration state registers are grouped together to match the memory access granularity, and each group of configuration state registers has a corresponding scoreboard entry.

Abstract: Systems, apparatuses, and methods for using a scoreboard to track updates to configuration state registers are disclosed. A system includes one or more processing nodes, one or more memory devices, a plurality of configuration state registers, and a communication fabric coupled to the processing unit(s) and memory device(s). The system uses a scoreboard to track updates to the configuration state registers during run-time. Prior to a node going into a power-gated state, the system stores only those configuration state registers that have changed. This reduces the amount of data written to memory on each transition into power-gated state, and increases the amount of time the node can spend in the power-gated state. Also, configuration state registers are grouped together to match the memory access granularity, and each group of configuration state registers has a corresponding scoreboard entry.

Abstract: Systems, apparatuses, and methods for implementing a fractional pointer lookup table are disclosed. A system includes a fractional pointer lookup table and control logic coupled to the table. The control logic performs an access to the table with a numerator and a denominator, wherein the numerator and the denominator are integers. The control logic receives a result of the lookup, wherein the result is either a rounded-up value of a quotient of the numerator and denominator or a rounded-down value of the quotient. In one embodiment, the control logic provides a fractional pointer to the table with each access and receives a fractional pointer limit from the table. The control logic initializes the fractional pointer to zero, increments the fractional pointer after each access to the table, and resets the fractional pointer to zero when the fractional pointer reaches the fractional pointer limit.

Abstract: A system for implementing load balancing schemes includes one or more processing units, a memory, and a communication fabric with a plurality of switches coupled to the processing unit(s) and the memory. A switch of the fabric determines a first number of streams on a first input port that are targeting a first output port. The switch also determines a second number of requestors, from all input ports, that are targeting the first output port. Then, the switch calculates a throttle factor for the first input port by dividing the first number of streams by the second number of streams. The switch applies the throttle factor to regulate bandwidth on the first input port for requestors targeting the first output port. The switch also calculates throttle factors for the other ports and applies the throttle factors when regulating bandwidth on the other ports.

Abstract: Systems, apparatuses, and methods for dynamic buffer management in multi-client token flow control routers are disclosed. A system includes at least one or more processing units, a memory, and a communication fabric with a plurality of routers coupled to the processing unit(s) and the memory. A router servicing multiple active clients allocates a first number of tokens to each active client. The first number of tokens is less than a second number of tokens needed to saturate the bandwidth of each client to the router. The router also allocates a third number of tokens to a free pool, with tokens from the free pool being dynamically allocated to different clients. The third number of tokens is equal to the difference between the second number of tokens and the first number of tokens. An advantage of this approach is reducing the amount of buffer space needed at the router.

Abstract: Systems, apparatuses, and methods for implementing a fractional pointer lookup table are disclosed. A system includes a fractional pointer lookup table and control logic coupled to the table. The control logic performs an access to the table with a numerator and a denominator, wherein the numerator and the denominator are integers. The control logic receives a result of the lookup, wherein the result is either a rounded-up value of a quotient of the numerator and denominator or a rounded-down value of the quotient. In one embodiment, the control logic provides a fractional pointer to the table with each access and receives a fractional pointer limit from the table. The control logic initializes the fractional pointer to zero, increments the fractional pointer after each access to the table, and resets the fractional pointer to zero when the fractional pointer reaches the fractional pointer limit.

Abstract: Systems, apparatuses, and methods for implementing dynamic control of a multi-region fabric are disclosed. A system includes at least one or more processing units, one or more memory devices, and a communication fabric coupled to the processing unit(s) and memory device(s). The system partitions the fabric into multiple regions based on different traffic types and/or periodicities of the clients connected to the regions. For example, the system partitions the fabric into a stutter region for predictable, periodic clients and a non-stutter region for unpredictable, non-periodic clients. The system power-gates the entirety of the fabric in response to detecting a low activity condition. After power-gating the entirety of the fabric, the system periodically wakes up one or more stutter regions while keeping the other non-stutter regions in power-gated mode. Each stutter region monitors stutter client(s) for activity and processes any requests before going back into power-gated mode.