Abstract:
A method for managing I/O operations in a shared file system environment. The method includes receiving for each of a plurality of compute nodes, information associated with I/O accesses to a shared file system, and applications for executing the I/O accesses. The method includes creating application profiles, based, at least in part, on the received information. The method then includes determining execution priorities for the application, based, at least in part, on the created application profiles.
Abstract:
In one form, a memory includes a memory bank, a page buffer, and an access circuit. The memory bank has a plurality of rows and a plurality of columns with volatile memory cells at intersections of the plurality of row and the plurality of columns. The page buffer is coupled to the plurality of columns and stores contents of a selected one of the plurality of rows. The access circuit is responsive to an adjacent command and a row address to perform a predetermined operation on the row address, and to refresh first and second addresses adjacent to the row address. In another form, a memory controller is adapted to interface with such a memory to select either a normal command or an adjacent command based on a number of activate commands sent to the row in a predetermined time window.
Abstract:
A data processor includes a memory accessing agent and a memory controller. The memory accessing agent generates a plurality of accesses to a memory. The memory controller is coupled to the memory accessing agent and schedules the plurality of memory accesses in an order based on characteristics of the memory. The characteristics of the memory include a row cycle page time (tRCPAGE) indicative of an acceptable number of activate commands to a row in the memory within a predetermined time window.
Abstract:
A data transfer device includes a FIFO memory and a control unit which obtains a data amount of the FIFO memory to control the FIFO memory and outputs a selection signal corresponding to the obtained data amount of the FIFO memory. An output data generation unit generates output data including either one of the second output data and an interpolation data selected based on the selection signal, and a first output data stored in a frame memory.
Abstract:
A method maintaining a fixed QoS for a PCIe device accessed by multiple hosts includes; receiving commands from the hosts in PCIe function queues of the PCIe device, fetching the commands from the PCIe function command queues, queuing the commands according to a command arbitration policy established for the PCIe device, storing the queued commands in an internal memory of the PCIe device, retrieving the queued commands from the internal memory in a sequence determined by applying a calculated QoS to at least one of the queued commands, and allocating PCIe device resources based on payload information corresponding to each one of the retrieved commands.
Abstract:
A method for managing I/O operations in a shared file system environment. The method includes receiving for each of a plurality of compute nodes, information associated with I/O accesses to a shared file system, and applications for executing the I/O accesses. The method includes creating application profiles, based, at least in part, on the received information. The method then includes determining execution priorities for the application, based, at least in part, on the created application profiles.
Abstract:
Methods and systems are disclosed for adaptive scheduling queue control based upon page hit distance determinations. A threshold occupancy value is determined for a window of previous access requests to a memory and used to adaptively control a number of access requests stored in a scheduling queue buffer. For certain embodiments, a page hit distance (PHD) determination for each access request and historical page hit distance data is used to adjust the threshold occupancy value that determines the number (N) of access requests stored in the buffer prior to removing an access request and using it to access the memory. For each access request, the page hit distance represents the number of previously received access requests since the last access request to access the same page of memory. An average PHD can be determined over a number (M) of previous access requests and used to control the threshold occupancy value.
Abstract:
A bridge for coupling a requesting interconnect and a serving interconnect connected to a number of coherent units in a computer system includes N machines configured to handle requests from the requesting interconnect and for handling allocated responses from the serving interconnect, wherein each of the N machines has an allocated local count; a tracking entity configured to track priorities of the N machines for issuing received requests to the serving interconnect and for issuing received responses to the requesting interconnect based on the N local counts; a first arbiter configured to arbitrate the received requests to be issued to the serving interconnect based on the tracked priorities; and a second arbiter configured to arbitrate the received responses to be issued to the requesting interconnect based on the tracked priorities.
Abstract:
An information processing device, comprising: a memory; and one or more central processing units coupled to the memory and configured to: control accesses to a device based on requests from users, record a start time of each access to the device and an end time of the access to the device, determine a load state of the device based on an elapsed time period from the start time to the end time, and limit, based on the load state of the device, a number of threads for one of the users, the threads being concurrently executed to access the device based on access requests to the device from the one of the users.
Abstract:
System and methods are provided for dynamically managing a first-in/first-out (FIFO) command queue of a system controller. One or more commands are received into the command queue, a command being associated with a priority parameter. A current command first in line to be executed in the command queue is determined, the current command being associated with a first priority parameter. A second command associated with a second priority parameter is determined, the second priority parameter being largest among priority parameters associated with the one or more commands. A final priority parameter for the current command is computed based at least in part on the second priority parameter.