In an example implementation, a method is provided. The method includes receiving at a switch a packet having a tuple, wherein the tuple comprises a plurality of fields, wherein the switch operates using a protocol that compares each one of the plurality of fields to a table of a plurality of tables in series. The switch compares the plurality of fields of the tuple to a single cache table instead of the plurality of tables in series. When a match is found, a plurality of actions associated with the plurality of fields of the tuple is accumulated. The plurality of actions is applied to the packet.

In an example, a system includes a network port that receives a packet; a first memory; a second memory; and a packet analyzer coupled to the network port. The packet analyzer operates to divide the packet into multiple fragments, analyze each of the multiple fragments to determine whether the corresponding fragment has a first priority level or a second, lower, priority level, determine whether to store each of the multiple fragments in the first memory or the second memory based on the priority level determined for that fragment, store each fragment determined to have the first priority level in the first memory, and store each fragment determined to have the second priority level in the second memory. The network port, packet analyzer and the first memory, which may be a cache memory, may be embodied on a chip, and the second memory may be external to the chip.

Methods, devices, and systems for facilitation of deterministic management of a plurality of electronic message packets communicated to an application via a network from a plurality of message sources. The facilitation involves an electronic message packet from the network, determining data indicative of order the electronic message packet was received relative to previously received electronic message packets, and providing the order to the application.

In one embodiment, a method is provided. The method of this embodiment provides storing a packet header at a set of at least one page of memory allocated to storing packet headers, and storing the packet header and a packet payload at a location not in the set of at least one page of memory allocated to storing packet headers.

A reception buffer of a packet reception apparatus includes a plurality of storage addresses. A packet determination unit receives a packet from a plurality of lines including a main system and an auxiliary system. The packet determination unit obtains a storage address corresponding to a unique number assigned to the packet, and overwrites and stores data of the packet onto the storage address. A packet extraction/transmission unit extracts and transmits the data stored in the reception buffer.

A device that is configured to generate reports to send to a server comprises at least one processor configured to generate the reports. Upon loss of connection, generated reports are stored in a buffer in memory of the device, capable of storing k reports. The first generated report is stored in a first part of the buffer, a most recently generated report is stored in a second part of the buffer, while the remaining reports are stored in a third part of the buffer. When the third part is full, a sampling algorithm is used to select the k2 reports to store in the third part.

The disclosure describes a data enqueuing method. The method may include: receiving a to-be-enqueued data packet, dividing the data packet into several slices to obtain slice information of the slices, and marking a tail slice of the data packet with a tail slice identifier; enqueuing corresponding slice information according to an order of the slices in the data packet, and in a process of enqueuing the corresponding slice information, if a slice is marked with the tail slice identifier, determining that the slice is the tail slice of the data packet, and generating a first-type node; and determining whether a target queue is empty, and if the target queue is empty, writing slice information of the tail slice into the target queue, and updating a head pointer of a queue head list according to the first-type node.

In one embodiment, a method is provided. The method of this embodiment provides storing a packet header at a set of at least one page of memory allocated to storing packet headers, and storing the packet header and a packet payload at a location not in the set of at least one page of memory allocated to storing packet headers.

Some embodiments provide a method for processing a packet received by a managed forwarding element. The method performs a series of packet classification operations based on header values of the received packet. The packet classifications operations determine a next destination of the received packet. When the series of packet classification operations specifies to send the packet to a network service that performs payload transformations on the packet, the method (1) assigns a service operation identifier to the packet that identifies the service operations for the network service to perform on the packet, (2) sends the packet to the network service with the service operation identifier, and (3) stores a cache entry for processing subsequent packets without the series of packet classification operations. The cache entry includes the assigned service operation identifier. The network service uses the assigned service operation identifier to process packets without performing its own classification operations.

Packet processors may implement burst absorption for processing network packets. Network packets may be processed at a packet processor according to a packet processing rate. For network packets that do not exceed the packet processing rate, processing of the network packets commences without an intervening buffer to store packet metadata. For network packets that exceed the packet processing rate, control metadata may be stored in a buffer at the excessive processing rate separate from packet data until removed from the buffer for processing according to the packet processing rate. In some embodiments, pipeline bubbles may be inserted by delaying the removal of packet metadata from the burst absorption buffer in order to start a pipeline bubble through forwarding stages of a packet processor.