Methods, computer systems, and computer readable medium are described for low latency reads using cached deduplicated data, including: receiving a request to read data from a storage system; query, using a generated hash value associated with the request to read data, one or more deduplication tables that corresponds to the hash value; and responsive to determining that the one or more deduplication tables includes an entry that corresponds to the hash value, using a mapping contained in the entry to perform the requested to read data, wherein the mapping includes a pointer to a physical location where at least a portion of the data is stored.

A system comprises a processor including a CPU core, first and second memory caches, and a memory controller subsystem. The memory controller subsystem speculatively determines a hit or miss condition of a virtual address in the first memory cache and speculatively translates the virtual address to a physical address. Associated with the hit or miss condition and the physical address, the memory controller subsystem configures a status to a valid state. Responsive to receipt of a first indication from the CPU core that no program instructions associated with the virtual address are needed, the memory controller subsystem reconfigures the status to an invalid state and, responsive to receipt of a second indication from the CPU core that a program instruction associated with the virtual address is needed, the memory controller subsystem reconfigures the status back to a valid state.

A system comprises a processor including a CPU core, first and second memory caches, and a memory controller subsystem. The memory controller subsystem speculatively determines a hit or miss condition of a virtual address in the first memory cache and speculatively translates the virtual address to a physical address. Associated with the hit or miss condition and the physical address, the memory controller subsystem configures a status to a valid state. Responsive to receipt of a first indication from the CPU core that no program instructions associated with the virtual address are needed, the memory controller subsystem reconfigures the status to an invalid state and, responsive to receipt of a second indication from the CPU core that a program instruction associated with the virtual address is needed, the memory controller subsystem reconfigures the status back to a valid state.

A request to write a first data item associated with a first thread to a memory device is received. The memory device includes a first portion and a second portion. The first portion includes a cache that includes a first block to be utilized for data caching and a second block and a third block to be used for block compaction. The second block is associated with a high modification frequency and the third block is associated with a low modification frequency. In response to determining a first memory page in the first block is available for writing the first data item, the first data item is written to the first memory page. A determination is made that a memory page criterion associated with the first thread has been satisfied. In response to identifying each of a set of second memory pages associated with the first thread written to at least one of the second block or the third block, the data of first memory page and each of the set of second memory pages is copied to the second portion of the memory device. The first memory page is marked as invalid on the first block and each of the set of second memory pages associated with the first thread are marked as invalid on at least one of the second block or the third block.

A data caching circuit and method are provided. The circuit is configured to cache data for a feature map calculated by a neural network, wherein a size of a convolution kernel of the neural network is K*K data, and a window corresponding to the convolution kernel slides at a step of S in the feature map, where K is a positive integer and S is a positive integer, the circuit comprising: a cache comprising K caching units, each caching unit being configured to respectively store a plurality of rows of the feature map, the plurality of rows comprising a corresponding row in every K consecutive rows of the feature map.

A network interface controller (NIC) capable of efficient management of non-idempotent operations is provided. The NIC can be equipped with a network interface, storage management logic block, and an operation management logic block. During operation, the network interface can receive a request for an operation from a remote device. The storage management logic block can store, in a local data structure, outcome of operations executed by the NIC. The operation management logic block can determine whether the NIC has previously executed the operation. If the NIC has previously executed the operation, the operation management logic block can obtain an outcome of the operation from the data structure and generate a response comprising the obtained outcome for responding to the request.

Systems and methods are provided for passing data amongst a plurality of switches having a plurality of links attached between the plurality of switches. At a switch, a plurality of load signals are received from a plurality of neighboring switches. Each of the plurality of load signals are made up of a set of values indicative of a load at each of the plurality of neighboring switches providing the load signal. Each value within the set of values provides an indication for each link of the plurality of links attached thereto as to whether the link is busy or quiet. Based upon the plurality of load signals, an output link for routing a received packet is selected, and the received packet is routed via the selected output link.

A network interface controller (NIC) capable of hybrid message matching is provided. The NIC can be equipped with a host interface, a hardware endpoint, and an endpoint management logic block. The host interface can couple the NIC to a host device. The hardware endpoint can facilitate a point of communication for an application running on the host device. The endpoint management logic block can maintain a list for storing a message associated with an endpoint represented by the hardware endpoint. The endpoint management logic block can then determine whether the utilization of the list is higher than a threshold. If the utilization is higher than the threshold, the endpoint management logic block can set a state of the endpoint to indicate that the endpoint is software managed. The NIC thus can transfer the control of the endpoint from the hardware endpoint to a software process of the host device.

Data-driven intelligent networking systems and methods are provided. The system can accommodate dynamic traffic with fast, effective congestion control. The system can maintain state information of individual packet flows, which can be set up or released dynamically based on injected data. Each flow can be provided with a flow-specific input queue upon arriving at a switch. Packets of a respective flow can be acknowledged after reaching the egress point of the network, and the acknowledgement packets can be sent back to the ingress point of the flow along the same data path. As a result, each switch can obtain state information of each flow and perform flow control on a per-flow basis.

Methods and systems are provided to facilitate network egress fairness between applications. At an egress port of a network, an arbitrator can provide fairness-based traffic shaping to data associated with applications. The desired fairness-based traffic shaping can be provided based on bandwidth, traffic classes, or other parameters. Consequently, the egress link's bandwidth can be allocated with fairness among the applications.