摘要:
An overflow control unit stores, in a FIFO buffer, data generated by a processor. The overflow control unit sets a predetermined flag, upon detecting that a FIFO buffer is full or nearly full. The overflow control unit stores, in a saving buffer, data sent from the processor, while the flag is set. Thereafter, the overflow control unit notifies the processor, by an interrupt, of an effect that an available capacity of the FIFO buffer rises above a predetermined threshold.Upon receiving an interrupt, the processor transfers to the FIFO buffer data saved in the saving buffer. Upon a completion of transferring to the FIFO buffer all data saved in the saving buffer, the processor resets the flag. This allows the overflow control unit to again store in the FIFO buffer, data sent from the processor. The overflow control unit also monitors the volume of data stored in the saving buffer, and notifies the processor, by an interrupt, of an effect that the saving buffer is full. Upon receiving an interrupt, the processor expands an available capacity of the saving buffer.
摘要:
When the processor writes a command data string into an address of the command entry area, a corresponding command is created by a bus interface. If an address output by the processor corresponds to a distributed shared memory area, the bus interface creates a remote access command. The send controller constructs a message based on the command created by the bus interface. This message is sent either to an interconnection network, or to a receive controller. The receive controller receives the message and interprets it. An address output by the processor is detected by a cache area access unit, and the cache area in the memory is accessed. When the processor receives an interrupt request while waiting for a response message to a remote read request, an deadlock control unit detects an abnormal end of the remote read request that the remote read has ended in an error, and, controls the processor to process the interrupt request with priority.
摘要:
A parallel computer including a plurality of processing elements, each of processing elements comprising a flag address holding unit for temporarily holding an address of a send complete flag of a direct remote write message when the direct remote write message is sent to another processing element, and a flag update unit for exclusively updating a flag represented by the address held in the flag address holding unit when data indicated by the direct remote write message has been sent.
摘要:
A parallel computer system capable of arbitrarily selecting nodes participating in barrier synchronization while enabling an arbitrary number of node groups to independently execute a process requiring the barrier synchronization. A communication network for the parallel computer system includes a plurality of routing controllers. Each routing controller has a register for setting a predetermined number of receipts of barrier synchronization request messages from other routing controllers, a destination to which the barrier synchronization request message is transmitted, and a destination to which a barrier synchronization establishment message is transmitted. If a destination of transmission of the barrier synchronization request message is set when completing a predetermined number of receipts of the barrier synchronization request messages and receipts of a barrier synchronization request message from a self-node, the barrier synchronization request message is transmitted to that destination of transmission. If the transmission destination is not set, the barrier synchronization establishment message is transmitted to the set transmission destination of the barrier synchronization establishment message. When receiving the barrier synchronization establishment message, the barrier synchronization establishment message is also transmitted to the set transmission destination of the barrier synchronization establishment message.
摘要:
A relay apparatus includes a first data storage unit and a processor. The first data storage unit stores an output port identifier in connection with a combination of a port identifier and a communication type identifier. The processor obtains, when a frame is received, a first output port identifier corresponding to a combination of a port identifier of a port that received the frame and a communication type identifier included in the received frame by use of the first data storage unit to output the received frame to a port identified by the first output port identifier.
摘要:
There is provided a relay device including: a buffer configured to store a received frame; a discarding unit configured to discard the received frame, when a utilization amount of the buffer exceeds a first value set corresponding to a communication type of the received frame; a first calculating unit configured to calculate the utilization amount of the buffer for each communication type at least two points in time, and calculate an amount of change in the utilization amount of the buffer for each communication type; and a setting unit configured to calculate the first value for each communication type, based on the amount of change in the utilization amount, and set the first value in the discarding unit.
摘要:
In a relay device, a first memory stores correspondence information representing a correspondence relationship between a node and a port. A second memory stores information by which a port to suppress flooding of a frame is distinguishable. A relay part limits a port, which floods a frame addressed to a node of which information is not stored in the first memory, based on the information stored in the second memory.
摘要:
A packet processing apparatus includes a static pattern matcher comparing pattern information defining a packet to be filtered with a value regarding at least a part of a received packet, the pattern information being stored by a pattern information manager. A frequency calculator calculates the frequency of matching by the static pattern matcher. A dynamic pattern matcher matches the frequency and a preset comparison value and a processing determiner determines a processing on the received packet based upon the dynamic pattern match.
摘要:
A relay system includes a first relay apparatus connected to a node through a first line and a second relay apparatus connected to the node through a second line. The first line and the second line belong to the same link aggregation group. The first relay apparatus includes a first control unit. The first control unit notifies, before relaying a received frame, the second relay apparatus of a source address included in the received frame in the absence of first relation information related to the source address in the first storage unit upon receiving the received frame via a port connected to the first line. The second relay apparatus includes a second control unit. The second control unit stores, in the second storage unit, second relation information regarding a relationship between the source address notified by the first relay apparatus and an output port connected to the second line.
摘要:
A packet transfer controlling apparatus includes a depth storage unit that stores a pattern and a depth in association with each other. When accepting a setting of the pattern, the apparatus searches the depth storage unit by using the accepted pattern to obtain the depth stored in association with the pattern, thereby deriving the depth of the information to be extracted from the input packet. Upon accepting the input of the packet, information is extracted sequentially from the head of the packet. The apparatus determines whether the extracted information is positioned at a location deeper than the derived depth. When determining that the extracted information is not yet positioned at a location deeper than the derived depth, the apparatus continues extraction of information. When determining that the extracted information is positioned at a location deeper than the depth, the apparatus ends extraction of information.