scispace - formally typeset
Search or ask a question

Showing papers on "Fast packet switching published in 2012"


Patent
31 Aug 2012
TL;DR: In this article, the authors propose a method for routing and switching data packets from one or more incoming links to one or multiple outgoing links of a router, where the data packet is extracted from the switching memory, and transmitted along the assigned outgoing link.
Abstract: A method for routing and switching data packets from one or more incoming links to one or more outgoing links of a router. The method comprises receiving a data packet from the incoming link, assigning at least one outgoing link to the data packet based on the destination address of the data packet, and after the assigning operation, storing the data packet in a switching memory based on the assigned outgoing link. The data packet extracted from the switching memory, and transmitted along the assigned outgoing link. The router may include a network processing unit having one or more systolic array pipelines for performing the assigning operation.

266 citations


Patent
21 Jun 2012
TL;DR: In this article, a method of tracking network traffic anomalies in a computing system, comprises receiving an ingress network packet at a configurable logic device (CLD), associating a timestamp with the packet, identifying at least one anomaly based on the contents of the packet and storing the anomalous packet and the timestamp in a persistent memory.
Abstract: A method of tracking network traffic anomalies in a computing system, comprises receiving an ingress network packet at a configurable logic device (CLD), associating a timestamp with the packet, identifying at least one anomaly based on the contents of the packet, and storing the anomalous packet and the timestamp in a persistent memory.

193 citations


Patent
20 Aug 2012
TL;DR: In this article, a forwarding information base (FIB) engine comprising a data store comprising forwarding instructions, an IP packet classifier comprising a processor configured to classify a first incoming IP packet as an information centric networking (ICN) packet and a second incoming ICN packet as a non-ICN packet, is coupled to the FIB engine and the IP packet processor.
Abstract: An internet protocol (IP) content router, comprising a forwarding information base (FIB) engine comprising a data store comprising forwarding instructions, an IP packet classifier comprising a processor configured to classify a first incoming IP packet as an information centric networking (ICN) packet and a second incoming IP packet as a non-ICN packet, and an IP packet processor comprising a processor, wherein the IP packet processor is coupled to the FIB engine and to the IP packet classifier, wherein the IP packet processor is configured to intercept the ICN packet for further ICN processing in the IP content router, consult the FIB engine for forwarding instructions for the non-ICN packet, and forward the non-ICN packet according to the forwarding instructions.

192 citations


Journal ArticleDOI
TL;DR: This paper exploits the abundant parallelism and other desirable features provided by current field-programmable gate arrays (FPGAs), and proposes a decision-tree-based, 2-D multi-pipeline architecture for next-generation packet classification.
Abstract: Multi-field packet classification has evolved from traditional fixed 5-tuple matching to flexible matching with arbitrary combination of numerous packet header fields. For example, the recently proposed OpenFlow switching requires classifying each packet using up to 12-tuple packet header fields. It has become a great challenge to develop scalable solutions for next-generation packet classification that support higher throughput, larger rule sets and more packet header fields. This paper exploits the abundant parallelism and other desirable features provided by current field-programmable gate arrays (FPGAs), and proposes a decision-tree-based, 2-D multi-pipeline architecture for next-generation packet classification. We revisit the techniques for traditional 5-tuple packet classification and propose several optimization techniques for the state-of-the-art decision-tree-based algorithm. Given a set of 12-tuple rules, we develop a framework to partition the rule set into multiple subsets each of which is built into an optimized decision tree. A tree-to-pipeline mapping scheme is carefully designed to maximize the memory utilization while sustaining high throughput. The implementation results show that our architecture can store either 10K real-life 5-tuple rules or 1K synthetic 12-tuple rules in on-chip memory of a single state-of-the-art FPGA, and sustain 80 and 40 Gbps throughput for minimum size (40 bytes) packets, respectively.

136 citations


Patent
06 Jul 2012
TL;DR: In this paper, the authors propose a method that processes network data through a network by determining whether the packet is a known packet and forwarding the packet to a managed switching element in the set of managed edge switching elements for forwarding to the network host.
Abstract: Some embodiments provide a method that processes network data through a network. The method receives a packet destined for a network host associated with a logical datapath set implemented by a set of managed edge switching elements and a set of managed non-edge switching elements in the network. The method determines whether the packet is a known packet. When the packet is a known packet, the method forwards the packet to a managed switching element in the set of managed edge switching elements for forwarding to the network host. When the packet is not a known packet, the method forwards the packet to a managed switching element in the set of managed non-edge switching elements for further processing.

96 citations


Proceedings ArticleDOI
25 Mar 2012
TL;DR: An accelerated version of libpcap is presented which gives significant speedups for many existing packet processing applications, and it is proved that existing software packet processing systems can be made adequate for high speed links, provided the authors are careful in removing other bottlenecks not related to packet I/O1.
Abstract: Software packet forwarding has been used for a long time in general purpose operating systems. While interesting for prototyping or on slow links, it is not considered a viable solution at very high packet rates, where various sources of overhead (particularly, the packet I/O mechanisms) get in the way of achieving good performance.

81 citations


Patent
08 May 2012
TL;DR: In this paper, the authors present a method for operating a network processor that receives a first data packet in a stream of data packets and a set of receive-queues adapted to store receive data packets.
Abstract: According to embodiments of the invention, there is provided a method for operating a network processor. The network processor receiving a first data packet in a stream of data packets and a set of receive-queues adapted to store receive data packets. The network processor processing the first data packet by reading a flow identification in the first data packet; determining a quality of service for the first data packet; mapping the flow identification and the quality of service into an index for selecting a first receive-queue for routing the first data packet; and utilizing the index to route the first data packet to the first receive-queue.

75 citations


Patent
06 Jul 2012
TL;DR: In this paper, a flow-based network switching system includes a memory having a flow table and a packet processor coupled to the memory, and the packet processor includes a user-programmable flowbased rule storage that includes a plurality of flowbased rules.
Abstract: A flow-based network switching system includes a memory having a flow table and a packet processor coupled to the memory. The packet processor includes a user-programmable flow-based rule storage that includes a plurality of flow-based rules. A flow-based handler and session manager in the packet processor is operable to retrieve application layer metadata from a first packet received over a network, determine a first flow session associated with the first packet using the application layer metadata from the first packet and the flow table, and retrieve at least one of the plurality of flow-based rules from the programmable flow-based rule storage using the application layer metadata from the first packet. A flow-based rule processing engine in the packet processor is operable to apply the at least one flow-based rule to the first packet. Packets with applied flow-based rules are forwarded through the network.

59 citations


Journal ArticleDOI
TL;DR: The optimum packet size is found to depend on the protocol characteristics, the bit rate, and the BER, and its effect on the normalized packet latency and on energy consumption is shown.
Abstract: In this paper, we investigate the effect of packet size selection on the performance of media access control (MAC) protocols for underwater wireless sensor networks, namely, carrier sense multiple access (CSMA) and the distance-aware collision avoidance protocol (DACAP). Our comparative analysis, conducted via ns-2 simulations, considers scenarios with varying, nonzero bit error rate (BER) and interference. We investigate metrics such as throughput efficiency (the ratio between the delivered bit rate and the offered bit rate), end-to-end packet latency, measured “per meter” to allow for different sizes of deployment areas, and the energy consumed to correctly deliver an information bit to the network collection point. Our results show the dependence of these metrics on the packet size, indicating the existence of an optimum. The optimum packet size is found to depend on the protocol characteristics, the bit rate, and the BER. For each protocol and scenario considered, we determine the packet size that optimizes throughput performance, and we show its effect on the normalized packet latency and on energy consumption.

57 citations


Patent
08 Mar 2012
TL;DR: In this paper, the authors describe packet switching in a packet switched network, in which each packet is associated with at least one switch tag and a switch tag includes a tag portion with switching information and an extended tag portion embedded with additional switching information.
Abstract: Methods and apparatus for managing packets in a packet switched network include, in at least one aspect, a device including: an input to receive a packet from one of a plurality of network devices, the plurality of network devices configured to communicate with one or more others of the network devices using a plurality of packets; and circuitry configured to control packet switching of the plurality of packets, at least one packet including: a switch tag including a tag portion embedded with switching information and an extended tag portion embedded with additional switching information, the switching information and the additional switching information configured to control a switching pattern associated with the at least one packet.

54 citations


Patent
Matsuoka Naoki1
18 Apr 2012
TL;DR: In this paper, a packet conversion scheme is proposed to convert a packet including a broadcast/multicast MAC address from a virtual machine into a multicast packet by setting the subnet identifier obtained based on the first association information and the packet type identifier obtained in a field of the MAC address of the packet to be transmitted, and transmits the packet, which is obtained by being converted, to the different computer via the network.
Abstract: Upon obtaining a packet including a broadcast/multicast MAC address from a virtual machine, a packet conversion apparatus obtains, based on first association information, a subnet identifier corresponding to the MAC address of the obtained packet, also obtains, based on second association information, a packet type identifier corresponding to the MAC address of the obtained packet, converts the packet to be transmitted to a different computer via a network into a multicast packet by setting the subnet identifier obtained based on the first association information and the packet type identifier obtained based on the second association information in a field of the MAC address of the packet to be transmitted, and transmits the packet to be transmitted, which is obtained by being converted, to the different computer via the network.

Patent
07 Sep 2012
TL;DR: In this paper, the in-switch delay is defined as a difference between a receiving time corresponding to reception of a packet and a forwarding or transmittal time correspond to forwarding or transmitting of the packet.
Abstract: Devices and methods for synchronizing devices over a switched fabric. A switch receives a request packet from a device, transmits a completion packet to the device, determines an in-switch delay, and stores the in-switch delay. Another switch receives a packet from a first device, forwards the packet to a second device, determines an in-switch delay of the packet, and stores the in-switch delay. Storing of in-switch delays may include adding an in-switch delay to values in one or more transaction delay fields of a packet. Storing of in-switch delays may include storing the delays in a storage element of a switch. In-switch delay may be determined as a difference between a receiving time corresponding to reception of a packet and a forwarding or transmittal time corresponding to forwarding or transmitting of a packet.

Journal ArticleDOI
TL;DR: The developed control system not only achieved autonomous distributed signaling and routing but also has a function that can adjust wavelength resources for optical packet and circuit switching autonomously in each link at each node.
Abstract: In this paper, we describe an autonomous distributed control system that we have been developing for an optical packet and circuit integrated network, and we experimentally evaluate its performance. Colored (i.e., multi-wavelength) optical packet-switched links transfer both control signals for circuit switching (e.g., signaling and routing) and best-effort packet data. We successfully transmitted high-definition uncompressed real-time video signals on two lightpaths established by our control system without degradation of video quality, simultaneously with other optical packet data transferred on the same optical fibers. Our developed control system not only achieved autonomous distributed signaling and routing but also has a function that can adjust wavelength resources for optical packet and circuit switching autonomously in each link at each node. Controllers achieved lightpath establishment within approximately 360 ms and dynamic resource adjustment within approximately 454 ms, in the best possible case in our experimental setup.

Patent
26 Oct 2012
TL;DR: In this article, a packet switch receives a management packet sent for discovery on a data flow's path through a network and creates a test packet (310T) which looks like a packet belonging to the data flow of interest.
Abstract: A packet switch receives a management packet sent for discovery on a data flow's path through a network. The switch creates a test packet (310T) which looks like a packet belonging to the data flow of interest. The test packet is processed by the switch like the data flow of interest except that the test packet is not necessarily transmitted. The processing results may include the egress ports for the test packet, and if the ports are part of a LAG or ECMP group, the results provide the group and the group members associated with the egress ports. Firewall actions and other information can be included. The switch may forward the results to the originator of the management packet. The switch may insert the results into the management packet and forward the management packet on the ports determined as egress ports for the test packet. Other features are also provided.

Patent
04 May 2012
TL;DR: In this paper, a packet of a flow is received at a static first level packet distribution module and a group of servers is statically selected for the packet of the flow with the first level module.
Abstract: A method, in one or more network elements that are in communication between clients that transmit packets and servers, of distributing the packets among the servers which are to process the packets. Stickiness of flows to servers assigned to process them is provided. A packet of a flow is received at a static first level packet distribution module. A group of servers is statically selected for the packet of the flow with the first level module. State that assigns the packet of the flow to the selected group of servers is not used. The packet of the flow is distributed to a distributed stateful second level packet distribution system. A server of the selected group is statefully selected with the second level system by accessing state that assigns processing of packets of the flow to the selected server. The packet of the flow is distributed to the selected server.

Patent
Karl Georg Hampel1
31 Jan 2012
TL;DR: In this article, a method and apparatus for processing packet streams associated with stream-based transport connection to enable communication between a source host and one or more destination hosts is presented and evaluated.
Abstract: A method and apparatus for processing packet streams associated with stream-based transport connection to thereby enable communication between a source host and one or more destination hosts.

Journal ArticleDOI
TL;DR: The network described here will contribute to diversification of services, enhanced functional flexibility, and efficient energy consumption, which are included in the twelve design goals of Future Networks announced by ITU-T (International Telecommunication Union - Telecommunication Standardization Sector).
Abstract: This paper presents recent progress made in the development of an optical packet and circuit integrated network. From the viewpoint of end users, this is a single network that provides both high-speed, inexpensive services and deterministic-delay, low-data-loss services according to the users' usage scenario. From the viewpoint of network service providers, this network provides large switching capacity with low energy requirements, high flexibility, and efficient resource utilization with a simple control mechanism. The network we describe here will contribute to diversification of services, enhanced functional flexibility, and efficient energy consumption, which are included in the twelve design goals of Future Networks announced by ITU-T (International Telecommunication Union - Telecommunication Standardization Sector). We examine the waveband-based network architecture of the optical packet and circuit integrated network. Use of multi-wavelength optical packet increases the switch throughput while minimizing energy consumption. A rank accounting method provides a solution to the problem of inter-domain signaling for end-to-end lightpath establishment. Moving boundary control for packet and circuit services makes for efficient resource utilization. We also describe related advanced technologies such as waveband switching, elastic lightpaths, automatic locator numbering assignment, and biologically-inspired control of optical integrated network.

Proceedings ArticleDOI
TL;DR: This paper addresses joint transceiver and relay design for a wireless multiple-input multiple-output (MIMO) switching scheme that enables data exchange among multiple users and shows that the optimized MIMO switching scheme based on the proposed algorithms significantly outperforms existing approaches in the literature.
Abstract: This paper addresses joint transceiver and relay design for a wireless multiple-input-multiple-output (MIMO) switching scheme that enables data exchange among multiple users. Here, a multi-antenna relay linearly precodes the received (uplink) signals from multiple users before forwarding the signal in the downlink, where the purpose of precoding is to let each user receive its desired signal with interference from other users suppressed. The problem of optimizing the precoder based on various design criteria is typically non-convex and difficult to solve. The main contribution of this paper is a unified approach to solve the weighted sum mean square error (MSE) minimization and weighted sum rate maximization problems in MIMO switching. Specifically, an iterative algorithm is proposed for jointly optimizing the relay's precoder and the users' receive filters to minimize the weighted sum MSE. It is also shown that the weighted sum rate maximization problem can be reformulated as an iterated weighted sum MSE minimization problem and can therefore be solved similarly to the case of weighted sum MSE minimization. With properly chosen initial values, the proposed iterative algorithms are asymptotically optimal in both high and low signal-to-noise ratio (SNR) regimes for MIMO switching, either with or without self-interference cancellation (a.k.a., physical-layer network coding). Numerical results show that the optimized MIMO switching scheme based on the proposed algorithms significantly outperforms existing approaches in the literature.

Journal ArticleDOI
TL;DR: This letter presents a simplified expression for the average packet error rate (PER) for an uncoded packet transmission system with large packet length over slow Rayleigh fading channels, by assuming that the bit error rate has an exponential form.
Abstract: This letter presents a simplified expression for the average packet error rate (PER) for an uncoded packet transmission system with large packet length over slow Rayleigh fading channels, by assuming that the bit error rate (BER) has an exponential form. Then the throughput performance of an uncoded packet transmission system with automatic repeat request (ARQ) is investigated, and the optimal packet length is derived to maximize the throughput. The theoretical analysis is validated by simulations.

Patent
03 Dec 2012
TL;DR: In this paper, a network device receives a packet with a multicast nexthop identifier, and creates a mask that includes addresses of egress packet forwarding engines, of the network device, to which to provide the packet.
Abstract: A network device receives a packet with a multicast nexthop identifier, and creates a mask that includes addresses of egress packet forwarding engines, of the network device, to which to provide the packet. The network device divides the mask into two portions, generates two copies of the packet, provides a first portion of the mask in a first copy of the packet, and provides a second portion of the mask in a second copy of the packet. The network device also forwards the first copy of the packet to an address of a first egress packet forwarding engine provided in the first portion of the mask, and forwards the second copy of the packet to an address of a second egress packet forwarding engine provided in the second portion of the mask.

Book ChapterDOI
25 May 2012
TL;DR: The paper proposes substitution of packet loss parameterP pl in ITU-T E-Model by effective packet loss P plef incorporating network jitter, a jitter buffer size and a packet size as additional input parameters for E- model.
Abstract: The paper deals with an influence of network jitter on effective packet loss in dejitter buffer We analyze behavior of jitter buffers with and without packet reordering capability and quantify the additional packet loss caused by packets dropped in buffer on top of the measured network packet loss We propose substitution of packet loss parameterP pl in ITU-T E-Model by effective packet loss P plef incorporating network jitter, a jitter buffer size and a packet size as additional input parameters for E-Model

Patent
17 Aug 2012
TL;DR: In this paper, the authors propose a method for forwarding a packet at a first managed interconnection switching element in a managed network of interconnection switches to a second one in a second network.
Abstract: Some embodiments provide a novel method for forwarding a packet at a first managed interconnection switching element in a managed network of interconnection switching elements. The method receives a packet from a source machine on a first network segment in a first network. The packet identifies a destination machine located on a second network segment in a second, different network. The method removes a context tag that identifies the first network segment. The method uses the identification of the first network segment to determine a logical switching element that connects at least the first and second network segments. The method encapsulates the packet with a logical context identifier that identifies a logical port of the logical switching element. The logical port corresponds to the second network segment. The method transmits the twice-encapsulated packet towards a second managed interconnection switching element at the second network.

Journal ArticleDOI
TL;DR: This paper proposes an approach of predictive compensation for simultaneous network-induced delays and packet losses, different from the majority of existing NCS control methods, that addresses co-design of both network and controller.

Proceedings ArticleDOI
10 Jun 2012
TL;DR: The numerical results obtained by simulations demonstrate that the proposed mechanism is able to significantly improve the firewall performance in terms of cumulative packet processing time compared to SA-BSPL technique.
Abstract: This paper presents a mechanism to improve firewall packet filtering time through optimizing the order of security policy filtering fields for early packet rejection. The proposed mechanism is based on the optimization of the filtering fields order according to traffic statistics. Furthermore, the mechanism uses multilevel packet filtering, and in each level unwanted packets are rejected as early as possible. So, the proposed mechanism can be considered also as a device protection mechanism against denial of service (DoS) attacks targeting the default policy rule. In addition, early packet acceptance is done through using the splay tree data structure which changes dynamically according to traffic flows. So, repeated packets will have less memory accesses and therefore reducing the overall packets matching time. The proposed technique aims to overcome some of the performance limitations of the previous technique, named Self Adjusting Binary Search on Prefix Length [1] (SA-BSPL). The numerical results obtained by simulations demonstrate that the proposed mechanism is able to significantly improve the firewall performance in terms of cumulative packet processing time compared to SA-BSPL technique.

Patent
30 Jan 2012
TL;DR: In this paper, when the bandwidth of a flow having a function to respond to congestion notification surpasses a first threshold, a congestion notification is given to rewrite a value representing the congestion state of the network out of the packet headers of packet received by the apparatus.
Abstract: In a packet relay apparatus or bandwidth policing apparatus, when the bandwidth of a flow having a function to respond to a congestion notification surpasses a first threshold, a congestion notification is given to rewrite a value representing the congestion state of the network out of the packet headers of the packet received by the apparatus. The packet is dropped if the bandwidth of the flow surpasses a second threshold higher than the first threshold. One of the following, including packet drop, rewriting the value representing the priority level in the network out of the packet headers and altering the priority level in the apparatus is imposed on the packet when the bandwidth of the flows surpasses a second threshold higher than the first threshold.

Proceedings ArticleDOI
03 Jul 2012
TL;DR: Matlab simulations show that considering packet loss in choosing the best communication path has a significant impact on reducing the energy consumption of the network as well as increasing network throughput.
Abstract: In this paper we analyze the importance of packet loss consideration (PLC) within inter node communication of a WSN. We evaluate the link quality between network nodes based on the link packet loss and not on the received signal strength indicator (RSSI). Our assessment is based on in field measurements obtained using the IRIS Crossbow WSN. The measurements are used to show the impact of packet loss in a known routing protocol for WSNs, LEACH. We propose a network initialization phase where all nodes gather information about the link packet loss from all neighbors. Using this information each node will choose the attending cluster head (CH) based on the smallest packet loss rather than on the highest RSSI. Matlab simulations show that considering packet loss in choosing the best communication path has a significant impact on reducing the energy consumption of the network as well as increasing network throughput.

Patent
31 Aug 2012
TL;DR: In this paper, a first stage filter receives a data frame comprising data packets representing respective encapsulated fragments of a PDU and parses a header of one of the packets to locate a start field, and determines whether the start field indicates that the packet comprises a start fragment of the PDU.
Abstract: Systems and methods for performing preprocessing of received data packets, to reduce processing burdens and improve efficiency of communications terminal receivers, are provided A first stage filter receives a data frame comprising data packets representing respective encapsulated fragments of a PDU The first stage filter parses a header of one of the packets to locate a start field, and determines whether the start field indicates that the packet comprises a start fragment of the PDU When the start field indicates that the packet comprises the start fragment, the first stage filter determines, based on packet label information of the packet header, whether to forward the packet to a second stage filter or to discard the packet, and forwards or discards the packet accordingly When the start field indicates that the packet does not comprise the start fragment, the first stage filter forwards the packet to the second stage filter

Journal ArticleDOI
TL;DR: An optical label processor for the RF tone in-band optical label based on parallel and asynchronous processing allows the optical packet switch with an exponential increase of number of ports at the expense of limited increase in the latency and the complexity.
Abstract: We propose an RF tone in-band labeling technique that is able to support large-scale and low-latency optical packet switch. This approach is based on N in-band wavelengths, each carrying M radio frequency (RF) tones. The wavelengths and the tones have a binary value, and are able to encode 2N×M possible routing address. We develop an optical label processor for the RF tone in-band optical label based on parallel and asynchronous processing. It allows the optical packet switch with an exponential increase of number of ports at the expense of limited increase in the latency and the complexity. By using RF tone in-band labeling technique, we demonstrate error free (bit error rate <; 10-9) optical packet switching operation for both 160 Gb/s packets and 40 Gb/s packets. We further investigate the scalability, the latency and the optical power fluctuation tolerance of the proposed RF tone in-band labeling technique. We show that 30 label bits are able to be delivered using single in-band wavelength, and the label processor introduces an extra latency of less than 7 ns.

Patent
Min Zha1, Ting Zou1, Qian Zhou1, Jing Huang1
12 Nov 2012
TL;DR: In this paper, the authors present a method, a device, and a system for forwarding a packet, which includes: receiving an IPv6 packet sent by a gateway, and obtaining a flow label of the packet, where the flow label corresponds to a link identity of customer premise equipment in one-to-one correspondence.
Abstract: The embodiments of the present invention provide a method, a device, and a system for forwarding a packet. The method includes: receiving an IPv6 packet sent by a gateway, and obtaining a flow label of the IPv6 packet, where the flow label of the IPv6 packet is corresponding to a link identity of customer premise equipment in one-to-one correspondence; decapsulating the IPv6 packet; performing network address translation on a decapsulated packet according to the flow label; and forwarding the decapsulated packet according to a translated network address. When multiple links between the gateway and an AFTR are corresponding to the same tunnel for network address translation, in the case that IP addresses of terminal users are overlapping or are the same, different terminal users can be distinguished, so that a terminal user that sends a packet can receive feedback, thus improving customer satisfaction.

Patent
24 Feb 2012
TL;DR: In this article, a packet assembler determines an associated packet reassembly of data portions and enqueues an identifier for each data portion in an input queue corresponding to the packet reassemb associated with the data portion.
Abstract: Described embodiments provide for processing received data packets into packet reassemblies for transmission as output packets of a network processor. A packet assembler determines an associated packet reassembly of data portions and enqueues an identifier for each data portion in an input queue corresponding to the packet reassembly associated with the data portion. A state data entry corresponding to each packet reassembly identifies whether the packet reassembly is actively processed by the packet assembler. Iteratively, until an eligible data portion is selected, the packet assembler selects a given data portion from a non-empty input queue for processing and determines if the selected data portion corresponds to a reassembly that is actively processed. If the reassembly is active, the packet assembler sets the selected data portion as ineligible for selection. Otherwise, the packet assembler selects the data portion for processing and modifies the packet reassembly based on the selected data portion.