Hardware Assisted Recursive Packet Classification Module for IPv6 etworks ABSTRACT

Size: px
Start display at page:

Download "Hardware Assisted Recursive Packet Classification Module for IPv6 etworks ABSTRACT"

Transcription

1 Hardware Assisted Recursive Packet Classification Module for IPv6 etworks Shivvasangari Subramani Department of Computer Science and Electrical Engineering University of Maryland Baltimore County,USA ABSTRACT In the advanced computer networks, to implement various internet functions such as traffic policing, Quality of Service, firewall processing, and normal unicast and multicast forwarding we need to classify packets based on its multi-field header which has time complexity. It is estimated that approximately 14% of the processing load is spent on packet classification, therefore, designing a hardware assisted packet classification scheme not only can reduce the work load placed on the network processor, but also potentially speed up the entire classification process. We propose a multiple SIMD processors based architecture to take advantage of the parallel processing paradigm offered by the decomposition algorithm. Our design is composed of seven identical SIMD processors to simultaneously handle the multiple data chunks from the various IP header fields. We used a single thread C code to simulate the classification process that occurs on a network processor, along with additional logging for individual instruction execution time and frequency. We have analyzed our code for packet classifier using various cache structures and sizes. And also performance analysis of single threaded Vs multithreaded architecture of packet classifier has been done. 1. I TRODUCTIO Packet classification is the process of comparing an incoming packet flow with a set of pre-established flow characteristics (known as filters or rules) to determine its identity. In the current generation of Internet Protocol - IPv4, packet classification is primarily used in enhancing security, monitoring network, and differentiating quality of service. With the emerging of the new internet procotol - IPv6, packet classification has exp1anded its roles to include matching flow characteristics to various flow profiles(know as Robust Header Compression Profile, defined in RFC3095 and RFC 3096, profile serves as a basis for different compression schemes targeting at various packet content) to determine its best header compression method. At its core, packet classification is a multiple fields search problem involving finding the best matching filter based on exact or wildcard pattern. Since filters can contain overlapping properties, search results often yield multiple matching filters for a single flow, thus, filter priority is also added to help eliminating non-exclusive search results. In the context of Robust Header Compression used for IPv6 in a wireless network, Packet Classification combined with Header encoding & decoding and CRC comprise the three components that are most computationally intensive on a network processor[ref 2]. Our design goal is motivated by the idea of forming a three piece hardware linked modules, work in conjunction to reduce the computational overhead imposed on the network processor (see Fig 1 below). Although this paper focuses just on the packet classification module, our overall design model still remains intact, and Figure 1 illustrates the placement of the packet classification module relative to the overall structure. Our packet classification module design is derived from a class of algorithms categorized as decomposition methods. In comparison with the remaining classification algorithms including decision tree, exhaustive 1

2 search and tuple space, decomposition stands out as a natural candidate for a hardware implementation. Decomposition method focuses on breaking down multiple field search into independent searches on single field, then combine all the search results. This type of search is a natural fit for a modern parallel SIMD processing architecture. Upon further research, we narrowed down our choice to the Recursive Flow Classification(RFC) due to its efficiency and speed. RFC treats the packet classification problem as a X-bit string reduction problem, where a X-bit string must be reduced from its original S-bit length to a new T-bit length such that T << S [ref 1]. The S-bit represents the total bits for all the header fields in a packet, and the T-bit represents a set of matching filters found for the S-bit header fields. The unique T-bit pattern used to represent the set of matching filters for a packet is known as an equivalence class identifier(eqid)[ref 1]. RFC carries out this bit-reduction process through multiple phases, each phase combines eqids returned from previous lookup phases, then subsequently re-applies the same reduction method to yield more concise eqid classifications with total lesser bit length. The last phase of this successive merging and reduction process yields a final eqid that specifies the flow action for the packet. The basic idea of RFC is illustrated in Figure 2[ref 1]: We divide the remaining of this paper into four major sections: Architecture Model, Architecture Simulation, Data Analysis and Conclusion. In section two - architecture model, we dive into the details of our RFC hardware architecture, explaining its various components and individual construction. In section three- 2

3 architecture simulation, we describe the simulation methods used to experiment with our design, and parameters employed for data inputs. In section four - data analysis, we present our simulation results and our analysis on the performance impact of the design. Finally, We conclude in section five with key findings of our simulation results and discuss some of the limitations in our experiment and their implications. 2. ARCHITECTURE MODEL The RFC Packet Classification architecture, depicted in Figure 3, can be summarized into four major parts: Buffer Units, Dispatching Unit(aka Control Unit), SIMD units, and Cache & Memory Units. The Buffer Units, described in section 2.1, is used to store input/output data stream, these streams include the original packet header fields and the intermediate filter search results from different RFC phases. The Buffer Units feed the data streams to the Dispatching Unit. The Dispatching Unit, described in section 2.2, acts as a load balancer for all the SIMD units, its major function is to issue the same instructions to all SIMD units, as well as divide the header fields or eqids into multiple data chunks and distribute them. The SIMD units(described in section 2.3) are the center piece of our design, they perform the necessary RFC bit-reduction process and forwards their search results(eqids) to the Output Buffer Unit. The last group of our design is the Cache & Memory Units(described in section 2.4), they serve as a storage place for instructions and classification filters. 2.1 BUFFER U IT The RFC RFC Packet Classification Architecture encompasses two separate piece of buffer units: Input Data Buffer and Output Data Buffer. The primary use of the Input Data Buffer unit is to capture the incoming packet header fields and queue them if necessary in order to pass the data down to the Dispatching Unit. The Input Buffer size is targeted at 512K, it can approximately queue IPv6 packet headers(each IPv6 header contains 352-bit header fields). The Input Buffer Unit is the equivalent of the original 2 S data source for the initial phase 0 shown in Figure 2. For the Output Buffer, the same identical hardware is used but the capacity is reduced in half since each reduction phase will produce data with much shorter bit-length. The purpose of the Output Data Buffer is to provide an intermediate storage area for keeping search results produced from different RFC phases. Using the example from Figure 2, the Output Data Buffer will store the search results 2 64, 2 24 and 2 12 from phase 1, 2 and 3. Once the Dispatching Unit detects the current incoming eqid is generated from a final RFC bit-reduction phase, it redirects the output 2 T to the output data link of our hardware module (see Figure 3). 2.2 DISPTACHI G U IT The Dispatching Unit is the central coordinator for distributing data and issuing instructions. It takes both the original packet header fields from the Input Data Buffer as well as the intermediate search results from the Output Buffer, the two data streams represent the entire data feed for the SIMD architecture. Raw data streams are further divided into multiple chunks of data groups by the Dispatching Unit and send to SIMD unit for processing. The dispatching unit has the ability to detect the processing condition for each unit and stalls if necessary when all SIMD units reach their capacities. Instruction fetch is done through a linkage to the Instruction Cache Unit, the Dispatching Unit loads and issues the same instruction to all the SIMD unit in the same execution cycle. Instructions are issued in-order and must be completed in-order in our model. The Dispatching Unit also ensures the current packet header has been completely processed before loads the next packet header from the Input Buffer. The final phase of a bit-reduction is detected when there is only one eqid coming from the Output Buffer Unit, once this condition occurs, the Dispatching Unit redirects the 3

4 final output to the output data link of the RFC hardware module for the next stage of processing by the network processor. 2.3 SIMD U ITS The SIMD units are the main work engine in our design, they carry out the filter search operation for each RFC bit-reduction phase and sends their results to the Output Buffer unit. We used seven SIMD units to accommodate the seven header fields in an IPv6 Packet Header. Since each RFC phase reduces the output data length, it means there is less input to feed back into SIMDs units, therefore some of the SIMD units may become idle after several phases. We considered the possibility of allowing fetching the next packet header in the queue, however, this potentially complicates the matter due to the fact the Output Buffer Unit will need to be shared. In addition, the Dispatching unit also must discern eqids between different packet headers. An apparent solution for the problem is to introduce some additional bit identifier attached to the data chunks, effectively labeling each packet header, but this will increase the data length and introduce extra processing overhead of stripping them during filter search, therefore, we decided not to address this issue in this design. In our design, all the SIMDs units connect to the memory unit and the Output Buffer unit via a common memory bus. The memory bus is configured to be 64 bit-wide and operates at a frequency of 200mhz. It is capable of 2 transfers per cycle, which offers a total 3200 megabytes per second bandwidth. In addition, our design targets each SIMD processor as a 32-bit processor with a clock rate of 100 mhz, and capable of processing 4 bytes per cycle. Assuming each packet header is exactly 352 bits(44 bytes), this configuration provides a processing power of 63 million packet headers per second, with a total of 2800 mega bytes per second throughput, which is well under the capacity limit of the memory bus. The remaining memory bus bandwidth can be used for loading the necessary filter data set from the main memory unit. 4

5 2.4 CACHE A D MEMORY U ITS The RFC hardware module is equipped with two cache units and a main memory unit to provide for instruction buffering and data fetching. The two cache units consist of an instruction cache unit and a victim cache unit. Both units adopt the fully associative cache structure aimed at reducing miss rate. Since our instruction data is shared across all the SIMD units, we believe providing a common shared instruction cache is the preferred option in this setup. The instruction cache is also linked to a victim cache to add a second level of instruction buffering. The victim cache stores data entries purged from the instruction cache due to conflict or capacity misses. The support for adding a victim cache comes from the fact that RFC recursively uses the same routines in multiple phases, there is a high probability of recently purged instruction will be used again in the next phase. The intended configuration for the cache is to set the cache size to be 8k and 4k respectively for the i-cache and v-cache. The data transfer between the instruction cache unit and the Dispatching unit is done via a memory bus similar to the bus architecture used to link SIMDs and the buffer units, however, this bus link is enhanced to run at a frequency of 400mhz -doubling the previous bus speed. All data transfer from cache unit is facilitated through the Dispatching Unit, there are no direct interactions between the SIMD units and cache, this is a sharp contrast in comparison with the main memory unit setup, which is directly connected to all the SIMDs units via the shared memory bus. The memory unit is engineered this way because it stores all the necessary filter data set relevant to the search operations being performed inside each SIMD unit. Since filter data sets tends to be large - approximately 4 mega bytes in size after compressed[ref 1], thus having a direct connection will likely reduce overhead and improve transfer speed. In our initial design, we contemplated the idea of having individual memory unit attached to each SIMD unit, and splitting the large filter data set to distribute them among the individual memory units, however, this approach will increase the complexity of the Dispatching Unit, which will have the increased responsibility of managing memory exchanges between SIMD units, this approach is also likely to increase the cost for the design, thus we decided to use a shared memory model for this architecture. 3.1 RULES FRAMI G 3. ARCHITECTURE SIMULATIO As the rules have been framed based on various fields of the header, a short description about the IPv6 header has been given[ref 3] IPv6 HEADER Figure 4: IPv6 Header 5

6 The IPV6 header consists of 40 bytes as given below: 1. Version - version 6 (4-bit IP version). 2. Traffic class - packet priority (8-bits). Priority values: 0-7 Low priority; 8-15 High Priority 3. Flow label - QoS management (20 bits). 4. Payload length - payload length in bytes (16 bits). 5. Next header - Specifies the next encapsulated protocol. The values are compatible with those specified for the IPv4 protocol field (8 bits). 6. Hop limit It is equivalent to the time to live field of IPv4 (8 bits). 7. Source and destination addresses bits each. For classifying a packet based on the header, we need to determine the fields which we have to consider and their various values. For example as we are dealing with IPV6 packets Version field is going to be the same. Hop limit and Payload length are not going to do anything with packet classification. Hence there is no need for considering these fields. However Traffic class, Flow label, Next header and address fields influence the type of packet as per their respective definitions. The values we have considered for various fields of header for classifying packets have been enlisted in the table 1. Table 1: Rule Table Address Type Destination address Flow Label Protocol Traffic class Rule o./ Priority Multicast FF00 **** **** **** IntServ TCP Low 29 Unicast 4000 **** **** **** IntServ TCP Low 30 Site Local FEC0 **** **** **** IntServ TCP Low 31 Link local FE80 **** **** **** IntServ TCP Low 32 Multicast FF00 **** **** **** DiffServ TCP Low 25 Unicast 4000 **** **** **** DiffServ TCP Low 26 Site Local FEC0 **** **** **** DiffServ TCP Low 27 Link local FE80 **** **** **** DiffServ TCP Low 28 Multicast FF00 **** **** **** IntServ TCP High 13 Unicast 4000 **** **** **** IntServ TCP High 14 Site Local FEC0 **** **** **** IntServ TCP High 15 Link local FE80 **** **** **** IntServ TCP High 16 Multicast FF00 **** **** **** DiffServ TCP High 9 6

7 Unicast 4000 **** **** **** Site Local FEC0 **** **** **** Link local FE80 **** **** **** Multicast FF00 **** **** **** Unicast 4000 **** **** **** Site Local FEC0 **** **** **** Link local FE80 **** **** **** Multicast FF00 **** **** **** Unicast 4000 **** **** **** Site Local FEC0 **** **** **** Link local FE80 **** **** **** Multicast FF00 **** **** **** Unicast 4000 **** **** **** Site Local FEC0 **** **** **** Link local FE80 **** **** **** Multicast FF00 **** **** **** Unicast 4000 **** **** **** Site Local FEC0 **** **** **** Link local FE80 **** **** **** DiffServ TCP High 10 DiffServ TCP High 11 DiffServ TCP High 12 IntServ UDP Low 21 IntServ UDP Low 22 IntServ UDP Low 23 IntServ UDP Low 24 DiffServ UDP Low 17 DiffServ UDP Low 18 DiffServ UDP Low 19 DiffServ UDP Low 20 IntServ UDP High 5 IntServ UDP High 6 IntServ UDP High 7 IntServ UDP High 8 DiffServ UDP High 1 DiffServ UDP High 2 DiffServ UDP High 3 DiffServ UDP High SYSTEM ARCHITECTURE The task of packet classification is accomplished by mapping S bits of packet header to T bits of CLASS ID. This mapping involves three phases which has been discussed below PHASE 0 The various fields of packet header which are valid for packet classification are divided into chunks and supplied as input to the phase 0. For example, with the first two bytes of destination address the type of the packet can be identified. I.e. the packet can be classified in one of the four categories such as multicast, site 7

8 local, link local, and unicast. Hence first byte of destination address is supplied to chunk#1 and second byte of destination address is given to chunk#2. Similarly chunk#3 is supplied with bits of CLASS field; chunk#4 is supplied with flow label and chunk#5 with next header. Mapping of actual input to the Equivalent ID (Eq ID) is done as given in the figure PHASE 1 The output of first two chunks in phase 0 is given as input to chunk#6 to determine the address type of the packet. The output of other three chunks in phase 0 is given as input to chunk#7 of phase 1. The mapping of these inputs to the corresponding Eq IDs is done as given in the figure PHASE 2 This is the final stage of the classifier where the Eq IDs of chunks#6 and #7 are combined to produce the CLASS ID of 5 bits which identifies the type of packet. The packet header with the size of 40 bytes has been reduced to the CLASS ID of size 5 bits. Figure 5: Implementation of packet Classifier 8

9 3.3 CACHE SETUP For performance analysis it is necessary to have various environments. In order to analyze the performance of the packet classifier, we have created a cache program with three classes. First class represents an I-cache and second class represents V-cache. Third class represents an I-cache connected with V-cache. First and second class combined together works in the conventional way. I.e. if the instruction is found in the I-cache it is a hit, else it marks it as miss, and goes to victim cache. In the third class where I-cache is connected with V-cache, if the instruction is not found in I-cache then V- cache is checked. If it exists there, then it still counts as a hit, if it is not in V-cache, it counts as a miss. Whenever cache is full, the oldest entry is stored into the v-cache. We have analyzed our code for packet classifier for various cache sizes. V-cache is always adjusted to be half the size of I-cache and also it supports same replacement strategy as that of I-cache (FIFO & Random). 4. DATA A ALYSIS From the implementation of the packet classifier module with a C program simulation we were able to observe certain results which helped us to evaluate the performance of our proposed architecture. A single threaded program to perform this kind of packet classification was run on a Simple Scalar simulator and the total simulation time was found to be 112 seconds. A similar C program which basically accepts the packet header as input, separate them into different chunks of data according to the specifications and gives the output for Phase 0 level took a total simulation time of 83 seconds. These two steps of the program which segregates the incoming packet into different chunks of data and generating the EqID (an output of Phase 0) is the tedious phase of execution in this program. When implemented in a multi threaded program Phase 0 takes the maximum amount of time since it involves comparing the input header with the required parameters and generating the intermediate EqID. Since always in a multithreaded program the total execution time of the program is same as the execution time of the longest phase we can state that the execution time for multithreaded implementation of the same program will be approximately 83 seconds. But however the overhead of switching between different threads in multithreading brings down the performance of the entire system. So, although the execution time is less in this case the overall performance of the system is brought down. Our single thread and multi-thread timing information provided us with a clue as to how a RFC hardware scheme will perform in a SIMD environment. We can observe that the parallel executing of search among SIMD hardware is definitely beneficial to the RFC based packet classification. Since in each phase, a single thread program can only process one chunk of data at a time, and the total time is the aggregation of all the time spend on processing all the headers, it is drastically different than SIMD execution, where for each phase, the worst timing is the longest execution time occurs on one of the SIMD unit, while the other units waiting for the last executing SIMD to finish. So for each, the total execution time will only be the worst execution or longest running time on one particular unit. The amount of improvement will be based on different data chunks we have to split among the SIMD, this is similar to the pipe-lined stage effect, so we anticipate the time to be only 1/N of the originally non-simd implementation, assume we have N multiple header fields. 9

10 An analysis of the miss rate for various size and types of cache implementations is tabulated as follows: I-cache without v-cache (random) I-cache with v-cache (random) o. of entries Miss rate 64 entries 61% 128 entries 24% 256 entries 12% 512 entries 12% I-cache without v-cache (FIFO) o. of entries Miss rate 64 entries 41% 128 entries 37% 256 entries 12% 512 entries 12% o. of entries Miss rate 64 entries 33% 128 entries 12% 256 entries 12% 512 entries 12% I-cache with v-cache (FIFO) o. of entries Miss rate 64 entries 38% 128 entries 12% 256 entries 12% 512 entries 12% Figure 7: Graph to show the impact in miss rate as we increase cache size and varying the replacement policy 10

11 From the above tabulated data we find that the miss rate reduces as the we increase the cache size, the v-cache plays a less and less role in helping bring down the miss rate as the size of the cache approaches 4KB and 8KB, which is what we targeted as our size for the i-cache, our results indicates the v-cache probably plays a much lesser role than we originally anticipated. But in all likelihood, can be safely removed from the overall design without any impact to performance if we decide to choose a cache size around 4KB to 8KB. Furthermore, based on our observation, we can speculate that the RFC algorithm uses media size quantities of instructions and if we provide an I-cache with 4KB or 8KB size, it is sufficient to accommodate the cache needs, and the addition of the v-cache will not add any more benefits. However, if the addition of 4KB or 8KB cache will significantly increase the cost of the overall cost of our design, and we have to work with small amount of cache hardware, then the addition of the v-cache can definitely be justified and will for sure improve the cache miss rates. One interesting note from the experiment is that it shows the miss rate in the Phase 0 (initial stage) are much higher than that of the Phase 1 and 2, it effectively illustrates the fact that when the initial cache contents are empty, the phase 0 suffer more misses, but as we proceed to the subsequent phases, the cache contents get filled and there is significant improvement in the miss rate. 5. Conclusions Although this project has helped us find the performance improvement considering the various simulation results, this will only be a speculation since we didn t actually simulate the fully functioning version of the SIMD model. We did not take into consideration the cost involved in constructing a 7 SIMD unit. Practically this cost might be too high to construct such a unit. We did not try to implement with a large size filter table to apparently show the benefit of parallel processing. A pipelined implementation of this architectural design could potentially result in significant increase in performance. In sum, from the above data analysis, using SIMD approach for this type of packet classification in IPv6 networks is sure to benefit when compared with all other approaches and our hopes are so high that our proposed model will work well in a real world implementation scenario. References [1] Pankaj Gupta and Nick McKeown, "Algorithms for Packet Classification", IEEE Network Special Issue, March/April 2001, volume 15, no. 2, pp [2] Taylor, David E.; Herkersdorf, Andreas; Doring, Andreas; Dittmann and Gero, "Robust Header Compression (ROHC) in Next-Generation Network Processors", IEEE/ACM Transactions on Networking, August 2005, Volume 13, Issue 4, pp [3] RFC Internet Protocol, Version 6 (IPv6) Specification. 11

CS 268: Route Lookup and Packet Classification

CS 268: Route Lookup and Packet Classification Overview CS 268: Route Lookup and Packet Classification Packet Lookup Packet Classification Ion Stoica March 3, 24 istoica@cs.berkeley.edu 2 Lookup Problem Identify the output interface to forward an incoming

More information

Quality of Service Mechanism for MANET using Linux Semra Gulder, Mathieu Déziel

Quality of Service Mechanism for MANET using Linux Semra Gulder, Mathieu Déziel Quality of Service Mechanism for MANET using Linux Semra Gulder, Mathieu Déziel Semra.gulder@crc.ca, mathieu.deziel@crc.ca Abstract: This paper describes a QoS mechanism suitable for Mobile Ad Hoc Networks

More information

EECS 122: Introduction to Computer Networks Switch and Router Architectures. Today s Lecture

EECS 122: Introduction to Computer Networks Switch and Router Architectures. Today s Lecture EECS : Introduction to Computer Networks Switch and Router Architectures Computer Science Division Department of Electrical Engineering and Computer Sciences University of California, Berkeley Berkeley,

More information

Generic Architecture. EECS 122: Introduction to Computer Networks Switch and Router Architectures. Shared Memory (1 st Generation) Today s Lecture

Generic Architecture. EECS 122: Introduction to Computer Networks Switch and Router Architectures. Shared Memory (1 st Generation) Today s Lecture Generic Architecture EECS : Introduction to Computer Networks Switch and Router Architectures Computer Science Division Department of Electrical Engineering and Computer Sciences University of California,

More information

Quality of Service II

Quality of Service II Quality of Service II Patrick J. Stockreisser p.j.stockreisser@cs.cardiff.ac.uk Lecture Outline Common QoS Approaches Best Effort Integrated Services Differentiated Services Integrated Services Integrated

More information

Packet Classification Using Dynamically Generated Decision Trees

Packet Classification Using Dynamically Generated Decision Trees 1 Packet Classification Using Dynamically Generated Decision Trees Yu-Chieh Cheng, Pi-Chung Wang Abstract Binary Search on Levels (BSOL) is a decision-tree algorithm for packet classification with superior

More information

Principles. IP QoS DiffServ. Agenda. Principles. L74 - IP QoS Differentiated Services Model. L74 - IP QoS Differentiated Services Model

Principles. IP QoS DiffServ. Agenda. Principles. L74 - IP QoS Differentiated Services Model. L74 - IP QoS Differentiated Services Model Principles IP QoS DiffServ Differentiated Services Architecture DSCP, CAR Integrated Services Model does not scale well flow based traffic overhead (RSVP messages) routers must maintain state information

More information

A Simulation: Improving Throughput and Reducing PCI Bus Traffic by. Caching Server Requests using a Network Processor with Memory

A Simulation: Improving Throughput and Reducing PCI Bus Traffic by. Caching Server Requests using a Network Processor with Memory Shawn Koch Mark Doughty ELEC 525 4/23/02 A Simulation: Improving Throughput and Reducing PCI Bus Traffic by Caching Server Requests using a Network Processor with Memory 1 Motivation and Concept The goal

More information

Lecture 13. Quality of Service II CM0256

Lecture 13. Quality of Service II CM0256 Lecture 13 Quality of Service II CM0256 Types of QoS Best Effort Services Integrated Services -- resource reservation network resources are assigned according to the application QoS request and subject

More information

Basics (cont.) Characteristics of data communication technologies OSI-Model

Basics (cont.) Characteristics of data communication technologies OSI-Model 48 Basics (cont.) Characteristics of data communication technologies OSI-Model Topologies Packet switching / Circuit switching Medium Access Control (MAC) mechanisms Coding Quality of Service (QoS) 49

More information

Multimedia Networking

Multimedia Networking CMPT765/408 08-1 Multimedia Networking 1 Overview Multimedia Networking The note is mainly based on Chapter 7, Computer Networking, A Top-Down Approach Featuring the Internet (4th edition), by J.F. Kurose

More information

AN ASSOCIATIVE TERNARY CACHE FOR IP ROUTING. 1. Introduction. 2. Associative Cache Scheme

AN ASSOCIATIVE TERNARY CACHE FOR IP ROUTING. 1. Introduction. 2. Associative Cache Scheme AN ASSOCIATIVE TERNARY CACHE FOR IP ROUTING James J. Rooney 1 José G. Delgado-Frias 2 Douglas H. Summerville 1 1 Dept. of Electrical and Computer Engineering. 2 School of Electrical Engr. and Computer

More information

Configuring global CAR 73 Overview 73 Configuring aggregate CAR 73 Configuration procedure 73 Configuration example 73

Configuring global CAR 73 Overview 73 Configuring aggregate CAR 73 Configuration procedure 73 Configuration example 73 Contents QoS overview 1 Introduction to QoS 1 QoS service models 1 Best-effort service model 1 IntServ model 1 DiffServ model 2 QoS techniques overview 2 Deploying QoS in a network 2 QoS processing flow

More information

UNIT I (Two Marks Questions & Answers)

UNIT I (Two Marks Questions & Answers) UNIT I (Two Marks Questions & Answers) Discuss the different ways how instruction set architecture can be classified? Stack Architecture,Accumulator Architecture, Register-Memory Architecture,Register-

More information

Configuring priority marking 63 Priority marking overview 63 Configuring priority marking 63 Priority marking configuration example 64

Configuring priority marking 63 Priority marking overview 63 Configuring priority marking 63 Priority marking configuration example 64 Contents QoS overview 1 Introduction to QoS 1 QoS service models 1 Best-effort service model 1 IntServ model 1 DiffServ model 2 QoS techniques overview 2 Deploying QoS in a network 2 QoS processing flow

More information

Advanced Computer Networks

Advanced Computer Networks Advanced Computer Networks QoS in IP networks Prof. Andrzej Duda duda@imag.fr Contents QoS principles Traffic shaping leaky bucket token bucket Scheduling FIFO Fair queueing RED IntServ DiffServ http://duda.imag.fr

More information

Virtual Memory. Chapter 8

Virtual Memory. Chapter 8 Virtual Memory 1 Chapter 8 Characteristics of Paging and Segmentation Memory references are dynamically translated into physical addresses at run time E.g., process may be swapped in and out of main memory

More information

KUPF: 2-Phase Selection Model of Classification Records

KUPF: 2-Phase Selection Model of Classification Records KUPF: 2-Phase Selection Model of Classification Records KAKIUCHI Masatoshi Nara Institute of Science and Technology Background Many Internet services classify the data to be handled according to rules

More information

a process may be swapped in and out of main memory such that it occupies different regions

a process may be swapped in and out of main memory such that it occupies different regions Virtual Memory Characteristics of Paging and Segmentation A process may be broken up into pieces (pages or segments) that do not need to be located contiguously in main memory Memory references are dynamically

More information

INTRODUCTION OF IPV6. Ravikumar Naik 21/11/2011

INTRODUCTION OF IPV6. Ravikumar Naik 21/11/2011 INTRODUCTION OF IPV6 Ravikumar Naik 21/11/2011 Outline Why we need a new version of the IP protocol? IPv6 Basics IPv6 Addressing Why we need a new version of the IP protocol? Contemporary studies indicated

More information

Audio Streams Merging Over ALMI

Audio Streams Merging Over ALMI Audio Streams Merging Over ALMI Christopher J. Dunkle, Zhen Zhang, Sherlia Y. Shi, Zongming Fei Department of Computer Science University of Kentucky 301 Rose Street, 2 nd floor Lexington, KY 40506-0495,

More information

RECOMMENDATION ITU-R BS.776 * Format for user data channel of the digital audio interface **

RECOMMENDATION ITU-R BS.776 * Format for user data channel of the digital audio interface ** Rec. ITU-R BS.776 1 RECOMMENDATION ITU-R BS.776 * Format for user data channel of the digital audio interface ** The ITU Radiocommunication Assembly considering (1992) a) that there is a real need for

More information

Network-Adaptive Video Coding and Transmission

Network-Adaptive Video Coding and Transmission Header for SPIE use Network-Adaptive Video Coding and Transmission Kay Sripanidkulchai and Tsuhan Chen Department of Electrical and Computer Engineering, Carnegie Mellon University, Pittsburgh, PA 15213

More information

Internet Protocols (chapter 18)

Internet Protocols (chapter 18) Internet Protocols (chapter 18) CSE 3213 Fall 2011 Internetworking Terms 1 TCP/IP Concepts Connectionless Operation Internetworking involves connectionless operation at the level of the Internet Protocol

More information

Lecture 24: Scheduling and QoS

Lecture 24: Scheduling and QoS Lecture 24: Scheduling and QoS CSE 123: Computer Networks Alex C. Snoeren HW 4 due Wednesday Lecture 24 Overview Scheduling (Weighted) Fair Queuing Quality of Service basics Integrated Services Differentiated

More information

CS6303 Computer Architecture Regulation 2013 BE-Computer Science and Engineering III semester 2 MARKS

CS6303 Computer Architecture Regulation 2013 BE-Computer Science and Engineering III semester 2 MARKS CS6303 Computer Architecture Regulation 2013 BE-Computer Science and Engineering III semester 2 MARKS UNIT-I OVERVIEW & INSTRUCTIONS 1. What are the eight great ideas in computer architecture? The eight

More information

Internet Services & Protocols. Quality of Service Architecture

Internet Services & Protocols. Quality of Service Architecture Department of Computer Science Institute for System Architecture, Chair for Computer Networks Internet Services & Protocols Quality of Service Architecture Dr.-Ing. Stephan Groß Room: INF 3099 E-Mail:

More information

Lecture: Large Caches, Virtual Memory. Topics: cache innovations (Sections 2.4, B.4, B.5)

Lecture: Large Caches, Virtual Memory. Topics: cache innovations (Sections 2.4, B.4, B.5) Lecture: Large Caches, Virtual Memory Topics: cache innovations (Sections 2.4, B.4, B.5) 1 More Cache Basics caches are split as instruction and data; L2 and L3 are unified The /L2 hierarchy can be inclusive,

More information

Chapter 8 Virtual Memory

Chapter 8 Virtual Memory Operating Systems: Internals and Design Principles Chapter 8 Virtual Memory Seventh Edition William Stallings Operating Systems: Internals and Design Principles You re gonna need a bigger boat. Steven

More information

CCVP QOS Quick Reference Sheets

CCVP QOS Quick Reference Sheets Why You Need Quality of Service (QoS)...3 QoS Basics...5 QoS Deployment...6 QoS Components...6 CCVP QOS Quick Reference Sheets Basic QoS Configuration...11 Traffic Classification and Marking...15 Queuing...26

More information

ip rsvp reservation-host

ip rsvp reservation-host Quality of Service Commands ip rsvp reservation-host ip rsvp reservation-host To enable a router to simulate a host generating Resource Reservation Protocol (RSVP) RESV messages, use the ip rsvp reservation-host

More information

Multimedia Networking. Network Support for Multimedia Applications

Multimedia Networking. Network Support for Multimedia Applications Multimedia Networking Network Support for Multimedia Applications Protocols for Real Time Interactive Applications Differentiated Services (DiffServ) Per Connection Quality of Services Guarantees (IntServ)

More information

Real-Time Applications. Delay-adaptive: applications that can adjust their playback point (delay or advance over time).

Real-Time Applications. Delay-adaptive: applications that can adjust their playback point (delay or advance over time). Real-Time Applications Tolerant: can tolerate occasional loss of data. Intolerant: cannot tolerate such losses. Delay-adaptive: applications that can adjust their playback point (delay or advance over

More information

Scalable Packet Classification for IPv6 by Using Limited TCAMs

Scalable Packet Classification for IPv6 by Using Limited TCAMs Scalable Packet Classification for IPv6 by Using Limited TCAMs Chia-Tai Chan 1, Pi-Chung Wang 1,Shuo-ChengHu 2, Chung-Liang Lee 1,and Rong-Chang Chen 3 1 Telecommunication Laboratories, Chunghwa Telecom

More information

High-Speed Network Processors. EZchip Presentation - 1

High-Speed Network Processors. EZchip Presentation - 1 High-Speed Network Processors EZchip Presentation - 1 NP-1c Interfaces Switch Fabric 10GE / N x1ge or Switch Fabric or Lookup Tables Counters SDRAM/FCRAM 64 x166/175mhz SRAM DDR NBT CSIX c XGMII HiGig

More information

Announcements. ! Previous lecture. Caches. Inf3 Computer Architecture

Announcements. ! Previous lecture. Caches. Inf3 Computer Architecture Announcements! Previous lecture Caches Inf3 Computer Architecture - 2016-2017 1 Recap: Memory Hierarchy Issues! Block size: smallest unit that is managed at each level E.g., 64B for cache lines, 4KB for

More information

MEMORY MANAGEMENT/1 CS 409, FALL 2013

MEMORY MANAGEMENT/1 CS 409, FALL 2013 MEMORY MANAGEMENT Requirements: Relocation (to different memory areas) Protection (run time, usually implemented together with relocation) Sharing (and also protection) Logical organization Physical organization

More information

VXLAN Overview: Cisco Nexus 9000 Series Switches

VXLAN Overview: Cisco Nexus 9000 Series Switches White Paper VXLAN Overview: Cisco Nexus 9000 Series Switches What You Will Learn Traditional network segmentation has been provided by VLANs that are standardized under the IEEE 802.1Q group. VLANs provide

More information

Memory Hierarchy. Slides contents from:

Memory Hierarchy. Slides contents from: Memory Hierarchy Slides contents from: Hennessy & Patterson, 5ed Appendix B and Chapter 2 David Wentzlaff, ELE 475 Computer Architecture MJT, High Performance Computing, NPTEL Memory Performance Gap Memory

More information

QoS Services with Dynamic Packet State

QoS Services with Dynamic Packet State QoS Services with Dynamic Packet State Ion Stoica Carnegie Mellon University (joint work with Hui Zhang and Scott Shenker) Today s Internet Service: best-effort datagram delivery Architecture: stateless

More information

Fundamental Questions to Answer About Computer Networking, Jan 2009 Prof. Ying-Dar Lin,

Fundamental Questions to Answer About Computer Networking, Jan 2009 Prof. Ying-Dar Lin, Fundamental Questions to Answer About Computer Networking, Jan 2009 Prof. Ying-Dar Lin, ydlin@cs.nctu.edu.tw Chapter 1: Introduction 1. How does Internet scale to billions of hosts? (Describe what structure

More information

Introduction to Internetworking

Introduction to Internetworking Introduction to Internetworking Stefano Vissicchio UCL Computer Science COMP0023 Internetworking Goal: Connect many networks together into one Internet. Any computer can send to any other computer on any

More information

Lecture 14: Performance Architecture

Lecture 14: Performance Architecture Lecture 14: Performance Architecture Prof. Shervin Shirmohammadi SITE, University of Ottawa Prof. Shervin Shirmohammadi CEG 4185 14-1 Background Performance: levels for capacity, delay, and RMA. Performance

More information

CSE398: Network Systems Design

CSE398: Network Systems Design CSE398: Network Systems Design Instructor: Dr. Liang Cheng Department of Computer Science and Engineering P.C. Rossin College of Engineering & Applied Science Lehigh University March 14, 2005 Outline Classification

More information

DISTRIBUTED EMBEDDED ARCHITECTURES

DISTRIBUTED EMBEDDED ARCHITECTURES DISTRIBUTED EMBEDDED ARCHITECTURES A distributed embedded system can be organized in many different ways, but its basic units are the Processing Elements (PE) and the network as illustrated in Figure.

More information

Lecture 3. The Network Layer (cont d) Network Layer 1-1

Lecture 3. The Network Layer (cont d) Network Layer 1-1 Lecture 3 The Network Layer (cont d) Network Layer 1-1 Agenda The Network Layer (cont d) What is inside a router? Internet Protocol (IP) IPv4 fragmentation and addressing IP Address Classes and Subnets

More information

Page 1. Quality of Service. CS 268: Lecture 13. QoS: DiffServ and IntServ. Three Relevant Factors. Providing Better Service.

Page 1. Quality of Service. CS 268: Lecture 13. QoS: DiffServ and IntServ. Three Relevant Factors. Providing Better Service. Quality of Service CS 268: Lecture 3 QoS: DiffServ and IntServ Ion Stoica Computer Science Division Department of Electrical Engineering and Computer Sciences University of California, Berkeley Berkeley,

More information

Lecture: Cache Hierarchies. Topics: cache innovations (Sections B.1-B.3, 2.1)

Lecture: Cache Hierarchies. Topics: cache innovations (Sections B.1-B.3, 2.1) Lecture: Cache Hierarchies Topics: cache innovations (Sections B.1-B.3, 2.1) 1 Types of Cache Misses Compulsory misses: happens the first time a memory word is accessed the misses for an infinite cache

More information

MA5400 IP Video Gateway. Introduction. Summary of Features

MA5400 IP Video Gateway. Introduction. Summary of Features MA5400 IP Video Gateway Introduction The MA5400 IP Video Gateway bridges the gap between the MPEG-2 and IP domains providing an innovative solution to the need to transport real-time broadcast quality

More information

Multiprocessors and Thread-Level Parallelism. Department of Electrical & Electronics Engineering, Amrita School of Engineering

Multiprocessors and Thread-Level Parallelism. Department of Electrical & Electronics Engineering, Amrita School of Engineering Multiprocessors and Thread-Level Parallelism Multithreading Increasing performance by ILP has the great advantage that it is reasonable transparent to the programmer, ILP can be quite limited or hard to

More information

Cache Performance! ! Memory system and processor performance:! ! Improving memory hierarchy performance:! CPU time = IC x CPI x Clock time

Cache Performance! ! Memory system and processor performance:! ! Improving memory hierarchy performance:! CPU time = IC x CPI x Clock time Cache Performance!! Memory system and processor performance:! CPU time = IC x CPI x Clock time CPU performance eqn. CPI = CPI ld/st x IC ld/st IC + CPI others x IC others IC CPI ld/st = Pipeline time +

More information

Queuing Mechanisms. Overview. Objectives

Queuing Mechanisms. Overview. Objectives Queuing Mechanisms Overview Objectives This module describes the queuing mechanisms that can be used on output interfaces. It includes the following topics: Queuing Overview FIFO Queuing Priority Queuing

More information

Tag Switching. Background. Tag-Switching Architecture. Forwarding Component CHAPTER

Tag Switching. Background. Tag-Switching Architecture. Forwarding Component CHAPTER CHAPTER 23 Tag Switching Background Rapid changes in the type (and quantity) of traffic handled by the Internet and the explosion in the number of Internet users is putting an unprecedented strain on the

More information

DiffServ Architecture: Impact of scheduling on QoS

DiffServ Architecture: Impact of scheduling on QoS DiffServ Architecture: Impact of scheduling on QoS Abstract: Scheduling is one of the most important components in providing a differentiated service at the routers. Due to the varying traffic characteristics

More information

Network Layer: Control/data plane, addressing, routers

Network Layer: Control/data plane, addressing, routers Network Layer: Control/data plane, addressing, routers CS 352, Lecture 10 http://www.cs.rutgers.edu/~sn624/352-s19 Srinivas Narayana (heavily adapted from slides by Prof. Badri Nath and the textbook authors)

More information

Tree-Based Minimization of TCAM Entries for Packet Classification

Tree-Based Minimization of TCAM Entries for Packet Classification Tree-Based Minimization of TCAM Entries for Packet Classification YanSunandMinSikKim School of Electrical Engineering and Computer Science Washington State University Pullman, Washington 99164-2752, U.S.A.

More information

Performance Improvement of Hardware-Based Packet Classification Algorithm

Performance Improvement of Hardware-Based Packet Classification Algorithm Performance Improvement of Hardware-Based Packet Classification Algorithm Yaw-Chung Chen 1, Pi-Chung Wang 2, Chun-Liang Lee 2, and Chia-Tai Chan 2 1 Department of Computer Science and Information Engineering,

More information

Lec 11 How to improve cache performance

Lec 11 How to improve cache performance Lec 11 How to improve cache performance How to Improve Cache Performance? AMAT = HitTime + MissRate MissPenalty 1. Reduce the time to hit in the cache.--4 small and simple caches, avoiding address translation,

More information

Quality of Service (QoS): Managing Bandwidth More Effectively

Quality of Service (QoS): Managing Bandwidth More Effectively 15 Quality of Service (QoS): Managing Bandwidth More Effectively Contents Introduction................................................. 15-2 Terminology............................................... 15-5

More information

TDT Coarse-Grained Multithreading. Review on ILP. Multi-threaded execution. Contents. Fine-Grained Multithreading

TDT Coarse-Grained Multithreading. Review on ILP. Multi-threaded execution. Contents. Fine-Grained Multithreading Review on ILP TDT 4260 Chap 5 TLP & Hierarchy What is ILP? Let the compiler find the ILP Advantages? Disadvantages? Let the HW find the ILP Advantages? Disadvantages? Contents Multi-threading Chap 3.5

More information

Slide Set 9. for ENCM 369 Winter 2018 Section 01. Steve Norman, PhD, PEng

Slide Set 9. for ENCM 369 Winter 2018 Section 01. Steve Norman, PhD, PEng Slide Set 9 for ENCM 369 Winter 2018 Section 01 Steve Norman, PhD, PEng Electrical & Computer Engineering Schulich School of Engineering University of Calgary March 2018 ENCM 369 Winter 2018 Section 01

More information

RMIT University. Data Communication and Net-Centric Computing COSC 1111/2061. Lecture 2. Internetworking IPv4, IPv6

RMIT University. Data Communication and Net-Centric Computing COSC 1111/2061. Lecture 2. Internetworking IPv4, IPv6 RMIT University Data Communication and Net-Centric Computing COSC 1111/2061 Internetworking IPv4, IPv6 Technology Slide 1 Lecture Overview During this lecture, we will understand The principles of Internetworking

More information

Routing Protocols in MANETs

Routing Protocols in MANETs Chapter 4 Routing Protocols in MANETs 4.1 Introduction The main aim of any Ad Hoc network routing protocol is to meet the challenges of the dynamically changing topology and establish a correct and an

More information

Managing and Securing Computer Networks. Guy Leduc. Chapter 2: Software-Defined Networks (SDN) Chapter 2. Chapter goals:

Managing and Securing Computer Networks. Guy Leduc. Chapter 2: Software-Defined Networks (SDN) Chapter 2. Chapter goals: Managing and Securing Computer Networks Guy Leduc Chapter 2: Software-Defined Networks (SDN) Mainly based on: Computer Networks and Internets, 6 th Edition Douglas E. Comer Pearson Education, 2015 (Chapter

More information

CSE 461 MIDTERM REVIEW

CSE 461 MIDTERM REVIEW CSE 461 MIDTERM REVIEW NETWORK LAYERS & ENCAPSULATION Application Application Transport Transport Network Network Data Link/ Physical Data Link/ Physical APPLICATION LAYER Application Application Used

More information

Lecture 3: Packet Forwarding

Lecture 3: Packet Forwarding Lecture 3: Packet Forwarding CSE 222A: Computer Communication Networks Alex C. Snoeren Thanks: Mike Freedman & Amin Vahdat Lecture 3 Overview Paper reviews Packet Forwarding IP Addressing Subnetting/CIDR

More information

6x86 PROCESSOR Superscalar, Superpipelined, Sixth-generation, x86 Compatible CPU

6x86 PROCESSOR Superscalar, Superpipelined, Sixth-generation, x86 Compatible CPU 1-6x86 PROCESSOR Superscalar, Superpipelined, Sixth-generation, x86 Compatible CPU Product Overview Introduction 1. ARCHITECTURE OVERVIEW The Cyrix 6x86 CPU is a leader in the sixth generation of high

More information

University of Toronto Faculty of Applied Science and Engineering

University of Toronto Faculty of Applied Science and Engineering Print: First Name:............ Solutions............ Last Name:............................. Student Number:............................................... University of Toronto Faculty of Applied Science

More information

H3C S9500 QoS Technology White Paper

H3C S9500 QoS Technology White Paper H3C Key words: QoS, quality of service Abstract: The Ethernet technology is widely applied currently. At present, Ethernet is the leading technology in various independent local area networks (LANs), and

More information

Chunyan Wang Electrical and Computer Engineering Dept. National University of Singapore

Chunyan Wang Electrical and Computer Engineering Dept. National University of Singapore Chunyan Wang Electrical and Computer Engineering Dept. engp9598@nus.edu.sg A Framework of Integrating Network QoS and End System QoS Chen Khong Tham Electrical and Computer Engineering Dept. eletck@nus.edu.sg

More information

Computer Networking: A Top Down Approach Featuring the. Computer Networks with Internet Technology, William

Computer Networking: A Top Down Approach Featuring the. Computer Networks with Internet Technology, William Dr. John Keeney 3BA33 TCP/IP protocol architecture with IP OSI Model Layers TCP/IP Protocol Architecture Layers TCP/IP Protocol Suite Application Layer Application Layer Telnet FTP HTTP DNS RIPng SNMP

More information

Interlaken Look-Aside Protocol Definition

Interlaken Look-Aside Protocol Definition Interlaken Look-Aside Protocol Definition Contents Terms and Conditions This document has been developed with input from a variety of companies, including members of the Interlaken Alliance, all of which

More information

Real-Time Protocol (RTP)

Real-Time Protocol (RTP) Real-Time Protocol (RTP) Provides standard packet format for real-time application Typically runs over UDP Specifies header fields below Payload Type: 7 bits, providing 128 possible different types of

More information

CS Computer Architecture

CS Computer Architecture CS 35101 Computer Architecture Section 600 Dr. Angela Guercio Fall 2010 Computer Systems Organization The CPU (Central Processing Unit) is the brain of the computer. Fetches instructions from main memory.

More information

Multicast and Quality of Service. Internet Technologies and Applications

Multicast and Quality of Service. Internet Technologies and Applications Multicast and Quality of Service Internet Technologies and Applications Aims and Contents Aims Introduce the multicast and the benefits it offers Explain quality of service and basic techniques for delivering

More information

Chapter 8 Virtual Memory

Chapter 8 Virtual Memory Operating Systems: Internals and Design Principles Chapter 8 Virtual Memory Seventh Edition William Stallings Modified by Rana Forsati for CSE 410 Outline Principle of locality Paging - Effect of page

More information

Network Control and Signalling

Network Control and Signalling Network Control and Signalling 1. Introduction 2. Fundamentals and design principles 3. Network architecture and topology 4. Network control and signalling 5. Network components 5.1 links 5.2 switches

More information

Quality of Service (QoS)

Quality of Service (QoS) Quality of Service (QoS) The Internet was originally designed for best-effort service without guarantee of predictable performance. Best-effort service is often sufficient for a traffic that is not sensitive

More information

RECOMMENDATION ITU-R BT.1720 *

RECOMMENDATION ITU-R BT.1720 * Rec. ITU-R BT.1720 1 RECOMMENDATION ITU-R BT.1720 * Quality of service ranking and measurement methods for digital video broadcasting services delivered over broadband Internet protocol networks (Question

More information

CS-435 spring semester Network Technology & Programming Laboratory. Stefanos Papadakis & Manolis Spanakis

CS-435 spring semester Network Technology & Programming Laboratory. Stefanos Papadakis & Manolis Spanakis CS-435 spring semester 2016 Network Technology & Programming Laboratory University of Crete Computer Science Department Stefanos Papadakis & Manolis Spanakis CS-435 Lecture #4 preview ICMP ARP DHCP NAT

More information

QUIZ: Longest Matching Prefix

QUIZ: Longest Matching Prefix QUIZ: Longest Matching Prefix A router has the following routing table: 10.50.42.0 /24 Send out on interface Z 10.50.20.0 /24 Send out on interface A 10.50.24.0 /22 Send out on interface B 10.50.20.0 /22

More information

An FPGA-Based Optical IOH Architecture for Embedded System

An FPGA-Based Optical IOH Architecture for Embedded System An FPGA-Based Optical IOH Architecture for Embedded System Saravana.S Assistant Professor, Bharath University, Chennai 600073, India Abstract Data traffic has tremendously increased and is still increasing

More information

Virtual Memory. Chapter 8

Virtual Memory. Chapter 8 Chapter 8 Virtual Memory What are common with paging and segmentation are that all memory addresses within a process are logical ones that can be dynamically translated into physical addresses at run time.

More information

QoS in IPv6. Madrid Global IPv6 Summit 2002 March Alberto López Toledo.

QoS in IPv6. Madrid Global IPv6 Summit 2002 March Alberto López Toledo. QoS in IPv6 Madrid Global IPv6 Summit 2002 March 2002 Alberto López Toledo alberto@dit.upm.es, alberto@dif.um.es Madrid Global IPv6 Summit What is Quality of Service? Quality: reliable delivery of data

More information

CS519: Computer Networks. Lecture 5, Part 5: Mar 31, 2004 Queuing and QoS

CS519: Computer Networks. Lecture 5, Part 5: Mar 31, 2004 Queuing and QoS : Computer Networks Lecture 5, Part 5: Mar 31, 2004 Queuing and QoS Ways to deal with congestion Host-centric versus router-centric Reservation-based versus feedback-based Window-based versus rate-based

More information

Routers Technologies & Evolution for High-Speed Networks

Routers Technologies & Evolution for High-Speed Networks Routers Technologies & Evolution for High-Speed Networks C. Pham Université de Pau et des Pays de l Adour http://www.univ-pau.fr/~cpham Congduc.Pham@univ-pau.fr Router Evolution slides from Nick McKeown,

More information

THE EFFICIENCY OF CONSTRAINT BASED ROUTING IN MPLS NETWORKS

THE EFFICIENCY OF CONSTRAINT BASED ROUTING IN MPLS NETWORKS VOLUME: 9 NUMBER: 5 SPECIAL ISSUE THE EFFICIENCY OF CONSTRAINT BASED ROUTING IN MPLS NETWORKS Martin MEDVECKY Department of Telecommunications, Faculty of Electrical Engineering and Information Technology,

More information

DEPARTMENT OF ELECTRONICS & COMMUNICATION ENGINEERING QUESTION BANK

DEPARTMENT OF ELECTRONICS & COMMUNICATION ENGINEERING QUESTION BANK DEPARTMENT OF ELECTRONICS & COMMUNICATION ENGINEERING QUESTION BANK SUBJECT : CS6303 / COMPUTER ARCHITECTURE SEM / YEAR : VI / III year B.E. Unit I OVERVIEW AND INSTRUCTIONS Part A Q.No Questions BT Level

More information

Virtual Memory - Overview. Programmers View. Virtual Physical. Virtual Physical. Program has its own virtual memory space.

Virtual Memory - Overview. Programmers View. Virtual Physical. Virtual Physical. Program has its own virtual memory space. Virtual Memory - Overview Programmers View Process runs in virtual (logical) space may be larger than physical. Paging can implement virtual. Which pages to have in? How much to allow each process? Program

More information

Scalable Packet Classification using Distributed Crossproducting of Field Labels

Scalable Packet Classification using Distributed Crossproducting of Field Labels Washington University in St. Louis Washington University Open Scholarship All Computer Science and Engineering Research Computer Science and Engineering Report Number: WUCSE-24-38 24-6-23 Scalable Packet

More information

Quality of Service in the Internet

Quality of Service in the Internet Quality of Service in the Internet Problem today: IP is packet switched, therefore no guarantees on a transmission is given (throughput, transmission delay, ): the Internet transmits data Best Effort But:

More information

Quality of Service in the Internet. QoS Parameters. Keeping the QoS. Leaky Bucket Algorithm

Quality of Service in the Internet. QoS Parameters. Keeping the QoS. Leaky Bucket Algorithm Quality of Service in the Internet Problem today: IP is packet switched, therefore no guarantees on a transmission is given (throughput, transmission delay, ): the Internet transmits data Best Effort But:

More information

CSE 461 Midterm Winter 2018

CSE 461 Midterm Winter 2018 CSE 461 Midterm Winter 2018 Your Name: UW Net ID: General Information This is a closed book/laptop examination. You have 50 minutes to answer as many questions as possible. The number in parentheses at

More information

Institute of Computer Technology - Vienna University of Technology. L73 - IP QoS Integrated Services Model. Integrated Services Model

Institute of Computer Technology - Vienna University of Technology. L73 - IP QoS Integrated Services Model. Integrated Services Model Integrated Services Model IP QoS IntServ Integrated Services Model Resource Reservation Protocol (RSVP) Agenda Integrated Services Principles Resource Reservation Protocol RSVP Message Formats RSVP in

More information

UNIT 2 TRANSPORT LAYER

UNIT 2 TRANSPORT LAYER Network, Transport and Application UNIT 2 TRANSPORT LAYER Structure Page No. 2.0 Introduction 34 2.1 Objective 34 2.2 Addressing 35 2.3 Reliable delivery 35 2.4 Flow control 38 2.5 Connection Management

More information

Quality of Service Monitoring and Delivery Part 01. ICT Technical Update Module

Quality of Service Monitoring and Delivery Part 01. ICT Technical Update Module Quality of Service Monitoring and Delivery Part 01 ICT Technical Update Module Presentation Outline Introduction to IP-QoS IntServ Architecture DiffServ Architecture Post Graduate Certificate in Professional

More information

Supra-linear Packet Processing Performance with Intel Multi-core Processors

Supra-linear Packet Processing Performance with Intel Multi-core Processors White Paper Dual-Core Intel Xeon Processor LV 2.0 GHz Communications and Networking Applications Supra-linear Packet Processing Performance with Intel Multi-core Processors 1 Executive Summary Advances

More information

On Distributed Communications, Rand Report RM-3420-PR, Paul Baran, August 1964

On Distributed Communications, Rand Report RM-3420-PR, Paul Baran, August 1964 The requirements for a future all-digital-data distributed network which provides common user service for a wide range of users having different requirements is considered. The use of a standard format

More information

Experimental Extensions to RSVP Remote Client and One-Pass Signalling

Experimental Extensions to RSVP Remote Client and One-Pass Signalling 1 Experimental Extensions to RSVP Remote Client and One-Pass Signalling Industrial Process and System Communications, Darmstadt University of Technology Merckstr. 25 D-64283 Darmstadt Germany Martin.Karsten@KOM.tu-darmstadt.de

More information

ANIC Host CPU Offload Features Overview An Overview of Features and Functions Available with ANIC Adapters

ANIC Host CPU Offload Features Overview An Overview of Features and Functions Available with ANIC Adapters ANIC Host CPU Offload Features Overview An Overview of Features and Functions Available with ANIC Adapters ANIC Adapters Accolade s ANIC line of FPGA-based adapters/nics help accelerate security and networking

More information