Network Layer Implemented Anycast Load Balancing

Size: px
Start display at page:

Download "Network Layer Implemented Anycast Load Balancing"

Transcription

1 Network Layer Implemented Anycast Load Balancing Ricardo Lopes Pereira IST, Inesc-ID Teresa Vazão IST, Inesc-ID Abstract This paper presents a server selection method where the server selection is performed by the network, at the first router from the client. This method, usable in networks under single administrative control, combines network metrics (such as traffic or latency) with common server load balancing metrics (such as load, CPU utilisation or response time), allowing network conditions to be taken into account when selecting the server which will handle the request. The performance of this technique was compared, by means of simulations, to that of widely used techniques such as random, round robin and least loaded server selection. 1 Introduction Despite the advent of the Personal Computers (PC) and distributed computing, the client/server model still plays a major role today. Many applications, such as Database servers and Web Services, remain centralised due to their nature. The ubiquity of the Internet or intranets, coupled with the emergence of the web browser as a human interface, also dictates that many applications are developed using the web server centric paradigm. However, this centralised approach has its weaknesses, as it presents a single point of failure. The concentration of all clients on the same server may also result in slow response times due to server overload, large network delays or congestion along the path from client to server. Replicating the application across several servers will eliminate the single point of failure. Server load can be mitigated by placing the application on a set of servers located in the same Local Area Network (LAN). Network latency and congestion may be avoided by placing servers across a Wide Area Network (WAN), closer to the clients. The two methods can be combined by spreading several clusters across a WAN. For each of these approaches, different load balancing methods are used to distribute the clients requests [1]. WAN load balancing techniques may take into account not only server availability but also the network distance and conditions along the paths from client to server and viceversa. This is also what routing protocols such as Open Shortest Path First (OSPF) do, compute shortest paths. Assigning each server the same anycast address [13] allows each router to know the path towards the nearest server. When clients use the anycast address to reach servers, the network layer will automatically forward packets to the nearest server (according to the routing metrics) [8]. This method is successfully being used today in the Internet to distribute User Datagram Protocol (UDP) queries by the Domain Name System (DNS) root servers [6]. This paper presents a transparent load balancing technique usable on networks under common administrative control, which aims at providing unmodified clients with better reply times while maximising the total number of requests handled by the set of servers. This method, which was named Network layer implemented Anycast Load Balancing (NALB), combines network proximity and server load information in order to achieve dynamic load balancing resorting to anycast addresses and link cost variation in already deployed routing protocols. This technique was validated by means of simulation using the HyperText Transfer Protocol (HTTP) protocol, although it may be applied to other applications such as DNS, Simple Mail Transfer Protocol (SMTP) or proxy servers. Section 2 details the anycast concept, its strengths and limitations. Section 3 presents the proposed load balancing scheme, which was compared with others by means of simulations, whose results are provided in section 4. The conclusions are presented in section 5. 2 IP Anycast The anycast concept was introduced into IP networking in [13], which defines it as: the best effort delivery of a datagram to at least one, and preferably only one, of the servers that accept datagrams for the anycast address. IPv6 is the first version of the IP protocol to support anycasting, albeit with limitations: anycast addresses may only be assigned to routers, not hosts, and may only be used as the destination

2 of a packet, never as the source [7]. Given several servers sharing the same anycast IP address and a proper routing infrastructure in place, clients should be directed to the nearest server (in number of hops or other routing metric). 2.1 Current combined anycast load balancing methods Different research proposals appeared combining anycast load balancing methods. Ticket based probing combines anycast with server load information by sending probes to several anycast servers. They reply indicating their status (load). The status information combined with the reply time provides server and network information which is used to select the server [3]. This method, however, is not transparent, requiring clients to be modified. Most importantly, it requires routers to be modified, for probe packets to reach several anycast servers. These demands complicate its deployment in a legacy network. Instead of leaving to the client the decision of which server to use, [1] proposes to restore it to the network, by using an active network. Active networks use routers capable of running user supplied code on each packet. In this approach, routers communicate with the servers, gathering load information, which may eventually be piggybacked on traffic from the servers. This information is then used to forward packets destined to the anycast server address. The router chooses the best server and changes the packet s target address from the anycast address to the best server s unicast address. When packets arrive in the opposite direction, the source address is changed, from the server s unicast address to the anycast one. This solution requires an active network, the development of a communication protocol between servers and routers and uses anycast addresses as the source of packets, which IPv6 disallows. A different approach consists in the utilisation of routing protocols which determine anycast routes using network metrics as well as server state information [9]. This method requires routers to collect and propagate server s state information. Legacy routers would have to be upgraded to support these routing protocols. 2.2 Anycast addressing and routing Anycast addresses can not be aggregated with other addresses as they may be dispersed by several subnetworks and their worldwide deployments would result in the explosion of the number of routing entries carried by each router. Nevertheless, they can be successfully used under single administrative control and several anycast intra-domain routing protocols have been proposed. Single-path routing anycast (SPRA) directs packets to the nearest server; it can be performed on small or confined networks using traditional routing protocols as OSPF [15]. Multi-path routing distributes packets among all servers within the same distance. Multi-path or global scale deployment of single-path routing requires specialised routing protocols [4, 17]. 2.3 Maintaining TCP connections Stateful connection methods, such as TCP, expect endpoints to remain the same throughout the life-time of the connection, but anycast routing does not ensure that packets from the same connection will always reach the same server. Routing tables may change due to topology changes or due to the introduction or removal of anycast servers. Several proposals address the lack of stateful transport support by anycast, either by adding support applications, modifying Transmission Control Protocol (TCP) or using IPv6 packet headers. A support application may be used to inform the client of the unicast address of an anycast server [17]. A UDP service, running at each anycast server on a well known port would receive UDP probe packets from the clients, which would use the anycast address as destination. The server would respond with a UDP message indicating its unicast address, which the client would then use to establish the TCP connection. Should several anycast servers respond, only the first one would be used. This method requires modifications to the clients and may not be used by legacy clients. Clients will also require a mechanism to know when a target address is an anycast address. This method adds a round-trip to the connection establishment process. A client could also determine the server s unicast address by sending a ping (ICMP ECHO request) to the anycast address. The server would then respond (ICMP ECHO reply) using its unicast address [11]. Modifications to the TCP protocol would enable it to overcome anycast limitations [13]. Just like in standard TCP, clients would send a SYN packet to the server s anycast address. However, the SYN-ACK response would arrive not from the anycast address but from the server s unicast address, which would be used from then on. A different proposal is to extend TCP s three-way handshake to a five-way handshake, creating a first phase where the unicast address of the server is determined [17]. These methods do not require the client to know, beforehand, when the address is an anycast one. Large scale TCP implementation substitution, albeit simpler than legacy applications modification, still represents a monumental endeavour. IPv6 packet extension headers may also be used to achieve the same goal [14]. The SYN-ACK response packet is extended with a header containing the anycast address, while it originates in the server s unicast address. The ex-

3 tension header allows the client to perceive the original address as anycast and allows it to match the SYN-ACK to its TCP connection. The ensuing packets will use the unicast address. This method requires a special TCP implementation. The IPv6 source route option has also been used to provide connection capabilities to anycast [5]. The client sends a normal SYN packet to the anycast address. The server responds with a SYN-ACK using its anycast address as source, but providing a route option header with its unicast address. The client may then, if and only if the packet is properly authenticated, reverse the route header, forwarding the next packets to the server s unicast address [7]. The server, upon receiving the packets at its unicast address, forwards them to the anycast address, according to the route header. The closest interface to the server with the anycast address will always belong to the server. Other than the added size, the use of the source route option will not overload the intermediate routers, as the route option header will only be processed at the hosts. Assuming the IPv6 source route option to be honoured at the TCP layer, this method would only require modifications to the servers software as well as an authentication infrastructure in place. However, this method infringes on IPv6 as the server sends packets originated in its anycast address. 2.4 IP Anycast evaluation In a network under single administrative control, current intra-domain routing protocols, such as OSPF, provide single-path anycast routing, by considering that paths to different servers sharing the same anycast address are different paths to the same server. Routing table dimension is not a problem as the number of anycast addresses used will be limited. This enable anycast use with stateless protocols such as DNS. The deployment of one of the methods described in section 2.3 on a single owner intranet is viable, enabling TCP connections. Strict standard compliance may also be overlooked if the anycast methods are confined to the intranet. In these conditions, SPRA load balancing is possible, directing clients to the nearest server, while also minimising network traffic. The possibility of transparently using anycast without requiring client modifications makes it an attractive option. However, [2] shows that scheduling each client to the nearest server (networkwise) may cause unbalanced servers, resulting in higher response times, due to large variations in each network region s traffic, attributable to population distribution, time zones and hour of day. The load unbalance caused by SPRA will prevent servers from being used to the fullest and will increase the response time for clients assigned to overloaded servers. A better result may be achieved by combining the network metrics used by the routing protocols with server status information. 3 Network layer implemented Anycast Load Balancing 3.1 NALB architecture NALB is an intra-domain architecture that provides load balancing by using a routing protocol that combines both network and server status information. NALB can also be combined with network QoS conditions should the intranet run a QoS enabled routing protocol [16]. Anycast addresses are assigned to the servers of a NALB domain and these servers will run the routing protocol, announcing a route to the anycast address. However, when receiving a packet for this address, a server will process it as his. The announced cost of the link to the anycast address will vary, according to the server state. Should the server be overloaded, the cost will be very high. Should it be idle, the cost would drop. When all the servers are idle, or equally loaded, clients will use their nearest server. When a server becomes overloaded, it will announce a higher cost to the anycast address, causing the routing protocol to consider it to be more distant. This will result in a number of clients now having a different nearest server, effectively being transfered from the overloaded server. Server failure is automatically supported as down servers will stop announcing their routes to the anycast address, being removed from the anycast group by the routing protocol. 3.2 Routing metrics Different types of metrics may be used to measure the server status, namely: CPU usage, run queue length, number of clients, transfer rate or a combination. However, should the load metric used be relative to each server s performance (like CPU usage) and not absolute (such as number of requests being served), the NALB method will automatically allow servers with different capacities to be used. The load metrics used, as well as the link cost variation, should be fine-tuned for best results experimentally. Furthermore, the link cost should only be increased when the load crosses a threshold, as assigning a client to a more distant server to avoid a mildly loaded server might prove a bad move. The periodicity of routing announcements will also have to be fine-tuned for best performance. A trade-off must be achieved between the added network traffic and router processing caused by frequent updating and the inaccurate view provided by making updates sparse. However, the added routing traffic may be compensated by the global traffic diminution granted by having clients use the nearest server.

4 3.3 Application examples Figure 1 presents a situation where two anycast servers provide service to four clients. Servers S1 and S2 each have their own unicast address but both share the anycast address A. Routers are connected by links with a unitary cost. In this situation, both servers are under low load, announcing a cost of one towards the anycast address. The dashed lines encompass the clients closest to each server: clients C1 and C2 will use server S1, clients C3 and C4 will use server S2. Figure 2 portrays a scenario where S1 is under heavier load than S2. S1 reacts by increasing the announced route cost towards the anycast address. As a result, C1 will now be closer to S2 and will use this server for future requests. S1 will reduce the number of potential clients to one whereas S2 will now have to handle three clients. Clients are diverted from the server with the highest load to the least loaded one. 4 Simulation results The proposed method was compared with traditional HTTP load balancing methods via simulation, performed using the 2.. version of the SSFNet network simulator [12]. The simulator was modified to implement CPU processing delays, traditional load balancing algorithms, anycast and the proposed method (including server performance measurement). The load balancing alternatives used were: SRPA, source IP address hashing, random, round robin and least loaded. The utilisation of a single server, with capacity equal to the sum of that of all servers, was also tested in order to provide a baseline for comparison. 4.1 Simulation scenario The core network of a large American ISP, with pointsof-presence in 25 cities, was used. It spans four time zones, Figure 1. Two servers with equal load Figure 2. Left server under heavier load making it a good candidate for a dynamic load balancing technique. To each of the 25 core routers, a network island was attached. Network islands are all identical, each having 2 HTTP clients connected at 1Mb/s, 3 at 1Mb/s and 4 at 1Mb/s. Servers, connected to the core routers using 1Mb/s links, were placed on opposite sides of the network. The OSPFv2 routing protocol was used. A stateful connection mechanism for anycast was assumed to be in place. HTTP 1.1 with persistent connections and no pipelining was used. Whenever a client established a connection, the scheduling algorithm was used by it to choose a server. Round robin state is kept globally. When a client chooses server 1, the next client will choose the second server and vice-versa. The least loaded algorithm is perfect, in the sense that it has instant up-to-date knowledge of the servers status. The least loaded server is considered to be the one with the smallest CPU run queue, to which clients have instant access with no network overhead. As the server selection is performed at the client, no dispatching overhead is included. Traditional dispatching mechanisms such as DNS, server redirection or triangulation will add further overhead to traditional load balancing techniques which does not affect SPRA or NALB. In these simulations, this overhead is not accounted for, providing traditional load balancing techniques with an advantage. The routing protocol uses link costs proportional to the link delay, which in turn is proportional to its distance. With NALB, servers generate new route announcements every 15 minutes. CPU utilisation is measured in 5 minute intervals. CPU s utilisation moving average is used as the server s load metric. The route to the anycast address will present a cost of 1 whenever this average is inferior to 5% and a cost from 1 to 25 when the moving average raises above this figure. The longest core links present a cost of 3. Four adjacent time zones were defined: zone has 6 islands, zone 1 has 2, zone 2 has 7 and zone 3 has 1. Simula-

5 tions were run over a 24 hour period, after a 2 hour warm-up period. The client generated traffic was defined as an office work day, with low traffic during the night and lunch time, and high activity during work hours (9 to 13 and 15 to 19). Clients are simulated to perform as real persons. They access the web site in sessions, which occur from time to time. During each session they will visit several pages, taking some time to read each one before moving to the next one. Each web page will be constituted by an HTML object and several binary objects such as images, each with its own size. For each object request the server will require some CPU time. Servers assign 5ms CPU time slices to each request using a round robin scheduler. Replies are only sent to the clients after the request has received the required CPU time. Clients timeout a request when the reply is not fully received within 2 minute. Simulations were performed with 1Kb/s, 1Mb/s and 1Mb/s links connecting the core routers. No traffic other than OSPF s and HTTP s existed. Links with 1Kb/s are insufficient to respond to the demands of the clients, becoming the bottleneck. In this case, the load balancing mechanism has to overcome the network limitations. Overdimensioned 1Mb/s links guarantee traffic to flow swiftly across the network, making servers the bottleneck. Here the load balancing mechanism has to divert clients to the least loaded server. 1Mb/s links provide a test where both server CPU and network capacity are at a premium. 4.2 Results analysis The existence of different timezones, each with a different number of clients, results in unbalanced traffic generation across the network. Fig. 3 shows how the load is distributed across both servers when using SPRA. The thin lines represent CPU usage while the thick lines represent the moving average. At the beginning of the day, clients in the east start peaking and, being closest to server 2, will cause it to have a higher load than server 1. Server 2 remains the most loaded one during the day as it is closer to a larger number of clients than server 1. Towards the end of the day, the roles are inverted. Server 2 s load starts to decrease as the end of the day approaches. On the other hand, server 1 still has a significant number of clients due to the 3 hour difference from the east to the west of the network, and becomes the most loaded server. On the whole, server 2 has to support a significantly larger load. Using NALB the load gap between servers is reduced as can be observed in Fig. 4. The thin lines at the bottom represent the link cost, which is read against the right vertical axis. At around 8h, server 2 crosses the CPU utilisation threshold, prompting the cost of its link to the anycast address to raise. As a result, the routing algorithm will determine that the shortest path to the anycast address for some CPU usage (%) H 8H S1 CPU S2 CPU 11H 14H Time (s) 17H 23H S1 Link cost S2 Link cost Figure 3. CPU usage using SPRA and 1Mb/s of the network islands now passes through server 1. Consequently, server 1 has an increase in traffic when compared to the previous SPRA scenario. Server 2 s traffic also increases more gradually. The same effect is visible during the afternoon period. This enables the set of servers to service a greater number of requests, as shown in table 1. NALB also fares well when compared to the classical load balancing methods. Table 1 shows the number of requests successfully satisfied by each method during the afternoon peak. Similar results were observed when considering the entire duration of the simulation. When the core network bandwidth is limited to only 1Kb/s, it becomes the bottleneck. Under this scenario,, server capacity is not an issue. NALB and SPRA behave exactly the same as CPU utilisation remains below 5%, preventing servers from announcing higher route costs. Anycast assigns each client to the closest server, limiting each request s traffic to part of the network. Client locality reduces global traffic, enabling a better utilisation of the available bandwidth. As a result, anycast is capable of answering a significantly larger number of requests. CPU usage (%) H 8H S1 CPU S2 CPU 11H 14H Time (s) 17H 23H S1 Link cost S2 Link cost Figure 4. CPU usage and link cost using NALB and 1Mb/s Link cost Link cost

6 Table 1. Successful requests from 14 to 17 Link speed Scheduler 1Kb/s 1Mb/s 1Mb/s NALB SPRA IP hash least loaded random round robin single server Reply time is also better, as shown in Fig. 5. For each load balancing method, the graphic shows the Cumulative Distribution Function (CDF) of the reply time. Due to the bandwidth limitations, no method is able to serve more than 2% of the requests in under 5 seconds. NALB and SPRA, whose lines overlap, show a clear advantage over the other protocols. IP hash, random and round robin present similar results as they all distribute clients over the two servers in almost uniform ways. The selection of the least loaded server will allocate most of the traffic to the first server, as server analysis is always performed in the same order, and run queue length will be zero. As a consequence, the network surrounding server 1 will be more loaded, decreasing its response capacity. Therefore, it will not perform as well as the other methods. The worst performance results from using a single server. As all packets must cross the core links surrounding the server, these will become overloaded. Nevertheless, as the server s CPU has double the capacity of all others, the processing time is shorter, resulting in a large number of successful replies under 1 second. When 1Mb/s links are used in the core, CPU also becomes scarce in the face of the large number of requests. In this situation, although SPRA performs the second best, it is outperformed by NALB. The SPRA method s performance is limited by the CPU shortage during peak hours. The NALB method is capable of serving a larger number of requests by diverting clients from the loaded server to the least loaded one. However, this comes at a price. Fig. 6 shows that the reply time using NALB is slightly higher than that of SPRA. The diverted clients get access to a server which provides faster response times but the path has higher delay. Network bandwidth is still a bottleneck, as can be observed by the relative performance of the other methods, which are similar to those attained using 1Kb/s links. 1Mb/s links allow the network not to obstacle performance. Server CPU speed becomes the single bottleneck. In these simulations, each server represents what, in a real deployment, would be a cluster of servers. Therefore, servers are quick to respond to requests, and are only overloaded during the peak hours of the day, as may be observed in Fig. 3 and 4. This results in very homogeneous results for the several methods used. Reply time is also very similar, as shown in Fig. 7. SPRA no longer performs significantly better than traditional methods with regard to the number of successful replies. NALB, however, by diverting further away clients to a least loaded server is capable of providing nearby clients with a fast response time, while only making far away clients move to the least loaded server. Clients further away from the loaded server may not be much further away from the new server, suffering little impact from the move, while releasing some capacity on the overloaded server. This explains the better response time of NALB. The Least loaded method performs comparably to NALB in terms of number of replies, but presents worse reply times, due to client to server paths. Round robin also performs very well as it provides the most uniform client distribution. Using a single server, with double the capacity, does not turn out to be a problem as the network is capable of forwarding all traffic easily..8.7 Cumulative request ratio.1 Cumulative request ratio NALB SPRA IP hash least loaded random round robin single server NALB SPRA.2 IP hash least loaded random.1 round robin single server Reply time (s) Reply time (s) Figure 5. Request reply time using 1Kb/s Figure 6. Request reply time using 1Mb/s

7 Cumulative request ratio NALB.3 SPRA IP hash.2 least loaded random.1 round robin single server Reply time (s) Figure 7. Request reply time using 1Mb/s 5 Conclusion In this paper a load balancing method that combines network proximity and server load information is presented. This is accomplished by combining anycast mechanisms with varying link costs, forcing the routing protocol to forward clients packets to the best server. This method was compared to ideal implementation of traditional load balancing methods and single path routing anycast using simulation. The results suggest that it performs as well as or better than the other methods under different network conditions. It also proved robust, providing consistent results under the different scenarios. Although this method is not yet applicable to the Internet at large, it could be deployed on large corporate or ISP networks. The method is completely distributed, supports servers with different capacities and can overcome server failure. The proposed load balancing method was used in HTTP load balancing simulations, yet nothing prevents it from being applied to other applications. Though simulations suggest this method to be better than traditional ones, further work could be performed to overcome some of its limitations. Anycast methods allow packets from client to server to travel the shortest path. However, in Web applications, the bulk of the traffic is in the opposite direction. A server which is more distant from the client might present a shorter path in the reverse direction. More complex server load metrics could be used, other than CPU usage. Running queue length, connected clients, bandwidth being consumed or a combination of these could be tested. Currently each server/cluster works independently, altering its link cost without regarding the other servers state. Cooperation between the servers could be established to better balance the load. The simulations were performed using a single server at each location. In a real deployment several servers would be used, in a cluster. Only one of them would run the routing protocol, requiring knowledge of the others state to produce an aggregate link cost. Resilience would have to be introduced into the protocol for this server to be replaced by one of the others should it fail. References [1] V. Cardellini, M. Colajanni, and P. S. Yu. Dynamic load balancing on web-server systems. 3(3):28 39, May-June [2] V. Cardellini, M. Colajanni, and P. S. Yu. Geographic load balancing for scalable distributed Web systems. In Proc. 8th International Symposium on Modeling, Analysis and Simulation of Computer and Telecommunication Systems, pages 2 27, San Francisco, CA, 2. [3] H. Chang, W. Jia, and L. Zhang. Distributed server selection with imprecise state for replicated server group. In Proc. 7th International Symposium on Parallel Architectures, Algorithms and Networks, pages 73 78, May 24. [4] S. Doi, S. Ata, H. Kitamura, and M. Murata. IPv6 anycast for simple and effective service-oriented communications. 42(5): , May 24. [5] R. Engel, V. Peris, and D. Saha. Using IP anycast for load distribution and server location. In Proc. 3rd Global Internet Mini-Conf., pages 27 35, Nov [6] T. Hardie. Distributing authoritative name servers via shared unicast addresses. IETF RFC3258, Apr. 22. [7] R. Hinden and S. Deering. Internet Protocol, Version 6 (IPv6) Specification. IETF RFC246, Dec [8] R. M. Hinden. IP next generation overview. Commun. ACM, 39(6):61 71, [9] C.-Y. Lin, J.-H. Lo, and S.-Y. Kuo. Load-balanced anycast routing. In Proc. Tenth International Conference on Parallel and Distributed Systems, pages 71 78, July 24. [1] H. Miura, M. Yamamoto, K. Nishimura, and H. Ikeda. Server load balancing with network support: Active anycast. In H. Yasuda, editor, IWAN, volume 1942 of Lecture Notes in Computer Science, pages Springer, 2. [11] M. Oe and S. Yamaguchi. Implementation and Evaluation of IPv6 Anycast. In Proc. 1th Annual Internet Soc. Conf., 2. [12] A. Ogielski, D. Nicol, J. Cowie, et al. SSFNet network simulator. [13] C. Partridge, T. Mendez, and W. Milliken. Host anycasting service. IETF RFC1546, Nov [14] S. Shah and D. Sanghi. Host Anycast Support in IPv6. In Proc. of 5th Int l Conf. on Advanced Computing (AD- COMP), Dec [15] K. H. Tan, M. Yaacob, T. C. Ling, and K. K. Phang. IPv6 single-path and multi-path anycast routing protocols performance analysis. In ISICT 3: Proceedings of the 1st international symposium on Information and communication technologies, pages Trinity College Dublin, 23. [16] Z. Wang. Internet QoS: Architectures and Mechanisms for Quality of Service. Morgan Kaufmann, Mar. 21. [17] S. Weber and L. Cheng. A survey of anycast in IPv6 networks. 42(1): , Jan. 24.

Load Balancing. Final Network Exam LSNAT. Sommaire. How works a "traditional" NAT? Un article de Le wiki des TPs RSM.

Load Balancing. Final Network Exam LSNAT. Sommaire. How works a traditional NAT? Un article de Le wiki des TPs RSM. Load Balancing Un article de Le wiki des TPs RSM. PC Final Network Exam Sommaire 1 LSNAT 1.1 Deployement of LSNAT in a globally unique address space (LS-NAT) 1.2 Operation of LSNAT in conjunction with

More information

Global Server Load Balancing

Global Server Load Balancing White Paper Overview Many enterprises attempt to scale Web and network capacity by deploying additional servers and increased infrastructure at a single location, but centralized architectures are subject

More information

IP Anycast: Point to (Any) Point Communications. Draft 0.3. Chris Metz, chmetz@cisco.com. Introduction

IP Anycast: Point to (Any) Point Communications. Draft 0.3. Chris Metz, chmetz@cisco.com. Introduction IP Anycast: Point to (Any) Point Communications Draft 0.3 Chris Metz, chmetz@cisco.com Introduction The Internet supports several different communication paradigms. Unicast is defined as a point-to-point

More information

FortiBalancer: Global Server Load Balancing WHITE PAPER

FortiBalancer: Global Server Load Balancing WHITE PAPER FortiBalancer: Global Server Load Balancing WHITE PAPER FORTINET FortiBalancer: Global Server Load Balancing PAGE 2 Introduction Scalability, high availability and performance are critical to the success

More information

Tunnel Broker System Using IPv4 Anycast

Tunnel Broker System Using IPv4 Anycast Tunnel Broker System Using IPv4 Anycast Xin Liu Department of Electronic Engineering Tsinghua Univ. lx@ns.6test.edu.cn Xing Li Department of Electronic Engineering Tsinghua Univ. xing@cernet.edu.cn ABSTRACT

More information

CS514: Intermediate Course in Computer Systems

CS514: Intermediate Course in Computer Systems : Intermediate Course in Computer Systems Lecture 7: Sept. 19, 2003 Load Balancing Options Sources Lots of graphics and product description courtesy F5 website (www.f5.com) I believe F5 is market leader

More information

Advanced Computer Networks. Layer-7-Switching and Loadbalancing

Advanced Computer Networks. Layer-7-Switching and Loadbalancing Oriana Riva, Department of Computer Science ETH Zürich Advanced Computer Networks 263-3501-00 Layer-7-Switching and Loadbalancing Patrick Stuedi, Qin Yin and Timothy Roscoe Spring Semester 2015 Outline

More information

Final Network Exam 01-02

Final Network Exam 01-02 1 ENSTB ITAM Final Network Exam 01-02 This exam is focused on Load balancing mechanisms. First part is related to "RFC 2391 : Load Sharing using IP Network Address Translation (LSNAT)" that was previously

More information

How To Provide Qos Based Routing In The Internet

How To Provide Qos Based Routing In The Internet CHAPTER 2 QoS ROUTING AND ITS ROLE IN QOS PARADIGM 22 QoS ROUTING AND ITS ROLE IN QOS PARADIGM 2.1 INTRODUCTION As the main emphasis of the present research work is on achieving QoS in routing, hence this

More information

Efficient DNS based Load Balancing for Bursty Web Application Traffic

Efficient DNS based Load Balancing for Bursty Web Application Traffic ISSN Volume 1, No.1, September October 2012 International Journal of Science the and Internet. Applied However, Information this trend leads Technology to sudden burst of Available Online at http://warse.org/pdfs/ijmcis01112012.pdf

More information

GLOBAL SERVER LOAD BALANCING WITH SERVERIRON

GLOBAL SERVER LOAD BALANCING WITH SERVERIRON APPLICATION NOTE GLOBAL SERVER LOAD BALANCING WITH SERVERIRON Growing Global Simply by connecting to the Internet, local businesses transform themselves into global ebusiness enterprises that span the

More information

Global Server Load Balancing

Global Server Load Balancing White Paper Global Server Load Balancing APV Series Application Delivery Controllers May 2011 Global Server Load Balancing Access. Security. Delivery. Introduction Scalability, high availability and performance

More information

LOAD BALANCING AS A STRATEGY LEARNING TASK

LOAD BALANCING AS A STRATEGY LEARNING TASK LOAD BALANCING AS A STRATEGY LEARNING TASK 1 K.KUNGUMARAJ, 2 T.RAVICHANDRAN 1 Research Scholar, Karpagam University, Coimbatore 21. 2 Principal, Hindusthan Institute of Technology, Coimbatore 32. ABSTRACT

More information

Understanding Slow Start

Understanding Slow Start Chapter 1 Load Balancing 57 Understanding Slow Start When you configure a NetScaler to use a metric-based LB method such as Least Connections, Least Response Time, Least Bandwidth, Least Packets, or Custom

More information

zhouzg281@nenu.edu.cn,{xugc,dengcy}@jlu.edu.cn,hejx@jlu.edu.cn,jianhuajiang@yahoo.com Anycast Member Server B Anycast Member Server A

zhouzg281@nenu.edu.cn,{xugc,dengcy}@jlu.edu.cn,hejx@jlu.edu.cn,jianhuajiang@yahoo.com Anycast Member Server B Anycast Member Server A ISBN 978-952-5726-00-8 (Print), 978-952-5726-01-5 (CD-ROM) Proceedings of the 2009 International Symposium on Web Information Systems and Applications (WISA 09) Nanchang, P. R. China, May 22-24, 2009,

More information

SiteCelerate white paper

SiteCelerate white paper SiteCelerate white paper Arahe Solutions SITECELERATE OVERVIEW As enterprises increases their investment in Web applications, Portal and websites and as usage of these applications increase, performance

More information

Avaya P330 Load Balancing Manager User Guide

Avaya P330 Load Balancing Manager User Guide Avaya P330 Load Balancing Manager User Guide March 2002 Avaya P330 Load Balancing Manager User Guide Copyright 2002 Avaya Inc. ALL RIGHTS RESERVED The products, specifications, and other technical information

More information

LOAD BALANCING MECHANISMS IN DATA CENTER NETWORKS

LOAD BALANCING MECHANISMS IN DATA CENTER NETWORKS LOAD BALANCING Load Balancing Mechanisms in Data Center Networks Load balancing vs. distributed rate limiting: an unifying framework for cloud control Load Balancing for Internet Distributed Services using

More information

Internet Protocol: IP packet headers. vendredi 18 octobre 13

Internet Protocol: IP packet headers. vendredi 18 octobre 13 Internet Protocol: IP packet headers 1 IPv4 header V L TOS Total Length Identification F Frag TTL Proto Checksum Options Source address Destination address Data (payload) Padding V: Version (IPv4 ; IPv6)

More information

Chapter 52 WAN Load Balancing

Chapter 52 WAN Load Balancing Chapter 52 WAN Load Balancing Introduction... 52-2 WAN Load Balancer Operating Principles... 52-2 Load Distribution Methods... 52-3 Round Robin Distribution... 52-3 Weighted Lottery Distribution... 52-3

More information

Internet Firewall CSIS 4222. Packet Filtering. Internet Firewall. Examples. Spring 2011 CSIS 4222. net15 1. Routers can implement packet filtering

Internet Firewall CSIS 4222. Packet Filtering. Internet Firewall. Examples. Spring 2011 CSIS 4222. net15 1. Routers can implement packet filtering Internet Firewall CSIS 4222 A combination of hardware and software that isolates an organization s internal network from the Internet at large Ch 27: Internet Routing Ch 30: Packet filtering & firewalls

More information

Network Level Multihoming and BGP Challenges

Network Level Multihoming and BGP Challenges Network Level Multihoming and BGP Challenges Li Jia Helsinki University of Technology jili@cc.hut.fi Abstract Multihoming has been traditionally employed by enterprises and ISPs to improve network connectivity.

More information

Scalability of web applications. CSCI 470: Web Science Keith Vertanen

Scalability of web applications. CSCI 470: Web Science Keith Vertanen Scalability of web applications CSCI 470: Web Science Keith Vertanen Scalability questions Overview What's important in order to build scalable web sites? High availability vs. load balancing Approaches

More information

Performance Evaluation of AODV, OLSR Routing Protocol in VOIP Over Ad Hoc

Performance Evaluation of AODV, OLSR Routing Protocol in VOIP Over Ad Hoc (International Journal of Computer Science & Management Studies) Vol. 17, Issue 01 Performance Evaluation of AODV, OLSR Routing Protocol in VOIP Over Ad Hoc Dr. Khalid Hamid Bilal Khartoum, Sudan dr.khalidbilal@hotmail.com

More information

AN EFFICIENT LOAD BALANCING ALGORITHM FOR A DISTRIBUTED COMPUTER SYSTEM. Dr. T.Ravichandran, B.E (ECE), M.E(CSE), Ph.D., MISTE.,

AN EFFICIENT LOAD BALANCING ALGORITHM FOR A DISTRIBUTED COMPUTER SYSTEM. Dr. T.Ravichandran, B.E (ECE), M.E(CSE), Ph.D., MISTE., AN EFFICIENT LOAD BALANCING ALGORITHM FOR A DISTRIBUTED COMPUTER SYSTEM K.Kungumaraj, M.Sc., B.L.I.S., M.Phil., Research Scholar, Principal, Karpagam University, Hindusthan Institute of Technology, Coimbatore

More information

Fault-Tolerant Framework for Load Balancing System

Fault-Tolerant Framework for Load Balancing System Fault-Tolerant Framework for Load Balancing System Y. K. LIU, L.M. CHENG, L.L.CHENG Department of Electronic Engineering City University of Hong Kong Tat Chee Avenue, Kowloon, Hong Kong SAR HONG KONG Abstract:

More information

A Task-Based Adaptive-TTL approach for Web Server Load Balancing *

A Task-Based Adaptive-TTL approach for Web Server Load Balancing * A Task-Based Adaptive-TTL approach for Web Server Load Balancing * Devarshi Chatterjee Zahir Tari RMIT University School of Computer Science and IT Melbourne, Australia zahirt@cs cs.rmit.edu.au * Supported

More information

Overview of TCP/IP. TCP/IP and Internet

Overview of TCP/IP. TCP/IP and Internet Overview of TCP/IP System Administrators and network administrators Why networking - communication Why TCP/IP Provides interoperable communications between all types of hardware and all kinds of operating

More information

8.2 The Internet Protocol

8.2 The Internet Protocol TCP/IP Protocol Suite HTTP SMTP DNS RTP Distributed applications Reliable stream service TCP UDP User datagram service Best-effort connectionless packet transfer Network Interface 1 IP Network Interface

More information

Using IPM to Measure Network Performance

Using IPM to Measure Network Performance CHAPTER 3 Using IPM to Measure Network Performance This chapter provides details on using IPM to measure latency, jitter, availability, packet loss, and errors. It includes the following sections: Measuring

More information

How To Balance A Web Server With Remaining Capacity

How To Balance A Web Server With Remaining Capacity Remaining Capacity Based Load Balancing Architecture for Heterogeneous Web Server System Tsang-Long Pao Dept. Computer Science and Engineering Tatung University Taipei, ROC Jian-Bo Chen Dept. Computer

More information

Chapter 51 WAN Load Balancing

Chapter 51 WAN Load Balancing Chapter 51 WAN Load Balancing Introduction... 51-2 WAN Load Balancer Operating Principles... 51-2 Load Distribution Methods... 51-3 Round Robin Distribution... 51-3 Weighted Lottery Distribution... 51-3

More information

Avaya ExpertNet Lite Assessment Tool

Avaya ExpertNet Lite Assessment Tool IP Telephony Contact Centers Mobility Services WHITE PAPER Avaya ExpertNet Lite Assessment Tool April 2005 avaya.com Table of Contents Overview... 1 Network Impact... 2 Network Paths... 2 Path Generation...

More information

Computer Networks - CS132/EECS148 - Spring 2013 ------------------------------------------------------------------------------

Computer Networks - CS132/EECS148 - Spring 2013 ------------------------------------------------------------------------------ Computer Networks - CS132/EECS148 - Spring 2013 Instructor: Karim El Defrawy Assignment 2 Deadline : April 25 th 9:30pm (hard and soft copies required) ------------------------------------------------------------------------------

More information

Chapter 7. Address Translation

Chapter 7. Address Translation Chapter 7. Address Translation This chapter describes NetDefendOS address translation capabilities. Dynamic Network Address Translation, page 204 NAT Pools, page 207 Static Address Translation, page 210

More information

Faculty of Engineering Computer Engineering Department Islamic University of Gaza 2012. Network Chapter# 19 INTERNETWORK OPERATION

Faculty of Engineering Computer Engineering Department Islamic University of Gaza 2012. Network Chapter# 19 INTERNETWORK OPERATION Faculty of Engineering Computer Engineering Department Islamic University of Gaza 2012 Network Chapter# 19 INTERNETWORK OPERATION Review Questions ٢ Network Chapter# 19 INTERNETWORK OPERATION 19.1 List

More information

Load balancing as a strategy learning task

Load balancing as a strategy learning task Scholarly Journal of Scientific Research and Essay (SJSRE) Vol. 1(2), pp. 30-34, April 2012 Available online at http:// www.scholarly-journals.com/sjsre ISSN 2315-6163 2012 Scholarly-Journals Review Load

More information

An Active Packet can be classified as

An Active Packet can be classified as Mobile Agents for Active Network Management By Rumeel Kazi and Patricia Morreale Stevens Institute of Technology Contact: rkazi,pat@ati.stevens-tech.edu Abstract-Traditionally, network management systems

More information

Content Distribution Networks (CDN)

Content Distribution Networks (CDN) 229 Content Distribution Networks (CDNs) A content distribution network can be viewed as a global web replication. main idea: each replica is located in a different geographic area, rather then in the

More information

CROSS LAYER BASED MULTIPATH ROUTING FOR LOAD BALANCING

CROSS LAYER BASED MULTIPATH ROUTING FOR LOAD BALANCING CHAPTER 6 CROSS LAYER BASED MULTIPATH ROUTING FOR LOAD BALANCING 6.1 INTRODUCTION The technical challenges in WMNs are load balancing, optimal routing, fairness, network auto-configuration and mobility

More information

MEASURING WORKLOAD PERFORMANCE IS THE INFRASTRUCTURE A PROBLEM?

MEASURING WORKLOAD PERFORMANCE IS THE INFRASTRUCTURE A PROBLEM? MEASURING WORKLOAD PERFORMANCE IS THE INFRASTRUCTURE A PROBLEM? Ashutosh Shinde Performance Architect ashutosh_shinde@hotmail.com Validating if the workload generated by the load generating tools is applied

More information

Load Balancing Mechanisms in Data Center Networks

Load Balancing Mechanisms in Data Center Networks Load Balancing Mechanisms in Data Center Networks Santosh Mahapatra Xin Yuan Department of Computer Science, Florida State University, Tallahassee, FL 33 {mahapatr,xyuan}@cs.fsu.edu Abstract We consider

More information

Performance of networks containing both MaxNet and SumNet links

Performance of networks containing both MaxNet and SumNet links Performance of networks containing both MaxNet and SumNet links Lachlan L. H. Andrew and Bartek P. Wydrowski Abstract Both MaxNet and SumNet are distributed congestion control architectures suitable for

More information

About Firewall Protection

About Firewall Protection 1. This guide describes how to configure basic firewall rules in the UTM to protect your network. The firewall then can provide secure, encrypted communications between your local network and a remote

More information

Web Caching and CDNs. Aditya Akella

Web Caching and CDNs. Aditya Akella Web Caching and CDNs Aditya Akella 1 Where can bottlenecks occur? First mile: client to its ISPs Last mile: server to its ISP Server: compute/memory limitations ISP interconnections/peerings: congestion

More information

Ensuring Business Continuity and Disaster Recovery with Coyote Point Systems Envoy

Ensuring Business Continuity and Disaster Recovery with Coyote Point Systems Envoy Ensuring Business Continuity and Disaster Recovery with Coyote Point Systems Envoy WHITE PAPER Prepared by: Lisa Phifer Core Competence, Inc. As e-business quickly becomes the norm, virtually every enterprise

More information

EECS 489 Winter 2010 Midterm Exam

EECS 489 Winter 2010 Midterm Exam EECS 489 Winter 2010 Midterm Exam Name: This is an open-book, open-resources exam. Explain or show your work for each question. Your grade will be severely deducted if you don t show your work, even if

More information

Network Architecture and Topology

Network Architecture and Topology 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 and routers 6. End systems 7. End-to-end

More information

Transport Layer Protocols

Transport Layer Protocols Transport Layer Protocols Version. Transport layer performs two main tasks for the application layer by using the network layer. It provides end to end communication between two applications, and implements

More information

Combining Global Load Balancing and Geo-location with Emissary TM

Combining Global Load Balancing and Geo-location with Emissary TM Combining Global Load Balancing and Geo-location with Emissary TM A New Kind of Global Internet Traffic Management Appliance from Coyote Point Systems and Digital Envoy Establishing a Geo-Sensitive, Highly

More information

Request Routing, Load-Balancing and Fault- Tolerance Solution - MediaDNS

Request Routing, Load-Balancing and Fault- Tolerance Solution - MediaDNS White paper Request Routing, Load-Balancing and Fault- Tolerance Solution - MediaDNS June 2001 Response in Global Environment Simply by connecting to the Internet, local businesses transform themselves

More information

The Effectiveness of Request Redirection on CDN Robustness

The Effectiveness of Request Redirection on CDN Robustness The Effectiveness of Request Redirection on CDN Robustness Limin Wang, Vivek Pai and Larry Peterson Presented by: Eric Leshay Ian McBride Kai Rasmussen 1 Outline! Introduction! Redirection Strategies!

More information

Avaya P333R-LB. Load Balancing Stackable Switch. Load Balancing Application Guide

Avaya P333R-LB. Load Balancing Stackable Switch. Load Balancing Application Guide Load Balancing Stackable Switch Load Balancing Application Guide May 2001 Table of Contents: Section 1: Introduction Section 2: Application 1 Server Load Balancing Section 3: Application 2 Firewall Load

More information

Reliable Multicast Protocol with Packet Forwarding in Wireless Internet

Reliable Multicast Protocol with Packet Forwarding in Wireless Internet Reliable Multicast Protocol with Packet Forwarding in Wireless Internet Taku NOGUCHI, Toru YOSHIKAWA and Miki YAMAMOTO College of Information Science and Engineering, Ritsumeikan University 1-1-1, Nojihigashi,

More information

Application Delivery Networking

Application Delivery Networking Application Delivery Networking. Raj Jain Washington University in Saint Louis Saint Louis, MO 63130 Jain@cse.wustl.edu These slides and audio/video recordings of this class lecture are at: 8-1 Overview

More information

Communications and Networking

Communications and Networking Communications and Networking History and Background telephone system local area networks Internet architecture: what the pieces are and how they fit together names and addresses: what's your name and

More information

Deploying in a Distributed Environment

Deploying in a Distributed Environment Deploying in a Distributed Environment Distributed enterprise networks have many remote locations, ranging from dozens to thousands of small offices. Typically, between 5 and 50 employees work at each

More information

Programming Assignments for Graduate Students using GENI

Programming Assignments for Graduate Students using GENI Programming Assignments for Graduate Students using GENI 1 Copyright c 2011 Purdue University Please direct comments regarding this document to fahmy@cs.purdue.edu. 1 OVERVIEW 2 1 Overview This document

More information

MANAGING NETWORK COMPONENTS USING SNMP

MANAGING NETWORK COMPONENTS USING SNMP MANAGING NETWORK COMPONENTS USING SNMP Abubucker Samsudeen Shaffi 1 Mohanned Al-Obaidy 2 Gulf College 1, 2 Sultanate of Oman. Email: abobacker.shaffi@gulfcollegeoman.com mohaned@gulfcollegeoman.com Abstract:

More information

Influence of Load Balancing on Quality of Real Time Data Transmission*

Influence of Load Balancing on Quality of Real Time Data Transmission* SERBIAN JOURNAL OF ELECTRICAL ENGINEERING Vol. 6, No. 3, December 2009, 515-524 UDK: 004.738.2 Influence of Load Balancing on Quality of Real Time Data Transmission* Nataša Maksić 1,a, Petar Knežević 2,

More information

Preparing Your IP Network for High Definition Video Conferencing

Preparing Your IP Network for High Definition Video Conferencing WHITE PAPER Preparing Your IP Network for High Definition Video Conferencing Contents Overview...3 Video Conferencing Bandwidth Demand...3 Bandwidth and QoS...3 Bridge (MCU) Bandwidth Demand...4 Available

More information

Lecture 3: Scaling by Load Balancing 1. Comments on reviews i. 2. Topic 1: Scalability a. QUESTION: What are problems? i. These papers look at

Lecture 3: Scaling by Load Balancing 1. Comments on reviews i. 2. Topic 1: Scalability a. QUESTION: What are problems? i. These papers look at Lecture 3: Scaling by Load Balancing 1. Comments on reviews i. 2. Topic 1: Scalability a. QUESTION: What are problems? i. These papers look at distributing load b. QUESTION: What is the context? i. How

More information

Technical Support Information Belkin internal use only

Technical Support Information Belkin internal use only The fundamentals of TCP/IP networking TCP/IP (Transmission Control Protocol / Internet Protocols) is a set of networking protocols that is used for communication on the Internet and on many other networks.

More information

International Journal of Scientific & Engineering Research, Volume 4, Issue 11, November-2013 349 ISSN 2229-5518

International Journal of Scientific & Engineering Research, Volume 4, Issue 11, November-2013 349 ISSN 2229-5518 International Journal of Scientific & Engineering Research, Volume 4, Issue 11, November-2013 349 Load Balancing Heterogeneous Request in DHT-based P2P Systems Mrs. Yogita A. Dalvi Dr. R. Shankar Mr. Atesh

More information

Path Selection Analysis in MPLS Network Based on QoS

Path Selection Analysis in MPLS Network Based on QoS Cumhuriyet Üniversitesi Fen Fakültesi Fen Bilimleri Dergisi (CFD), Cilt:36, No: 6 Özel Sayı (2015) ISSN: 1300-1949 Cumhuriyet University Faculty of Science Science Journal (CSJ), Vol. 36, No: 6 Special

More information

QoS issues in Voice over IP

QoS issues in Voice over IP COMP9333 Advance Computer Networks Mini Conference QoS issues in Voice over IP Student ID: 3058224 Student ID: 3043237 Student ID: 3036281 Student ID: 3025715 QoS issues in Voice over IP Abstract: This

More information

Superior Disaster Recovery with Radware s Global Server Load Balancing (GSLB) Solution

Superior Disaster Recovery with Radware s Global Server Load Balancing (GSLB) Solution Superior Disaster Recovery with Radware s Global Server Load Balancing (GSLB) Solution White Paper January 2012 Radware GSLB Solution White Paper Page 1 Table of Contents 1. EXECUTIVE SUMMARY... 3 2. GLOBAL

More information

APPENDIX 1 USER LEVEL IMPLEMENTATION OF PPATPAN IN LINUX SYSTEM

APPENDIX 1 USER LEVEL IMPLEMENTATION OF PPATPAN IN LINUX SYSTEM 152 APPENDIX 1 USER LEVEL IMPLEMENTATION OF PPATPAN IN LINUX SYSTEM A1.1 INTRODUCTION PPATPAN is implemented in a test bed with five Linux system arranged in a multihop topology. The system is implemented

More information

Efficient Parallel Distributed Load Balancing in Content Delivery Networks

Efficient Parallel Distributed Load Balancing in Content Delivery Networks Efficient Parallel Distributed Load Balancing in Content Delivery Networks P.Jyothi*1, N.Rajesh*2, K.Ramesh Babu*3 M.Tech Scholar, Dept of CSE, MRECW, Maisammaguda, Secunderabad, Telangana State, India,

More information

CS 457 Lecture 19 Global Internet - BGP. Fall 2011

CS 457 Lecture 19 Global Internet - BGP. Fall 2011 CS 457 Lecture 19 Global Internet - BGP Fall 2011 Decision Process Calculate degree of preference for each route in Adj-RIB-In as follows (apply following steps until one route is left): select route with

More information

EINDHOVEN UNIVERSITY OF TECHNOLOGY Department of Mathematics and Computer Science

EINDHOVEN UNIVERSITY OF TECHNOLOGY Department of Mathematics and Computer Science EINDHOVEN UNIVERSITY OF TECHNOLOGY Department of Mathematics and Computer Science Examination Computer Networks (2IC15) on Monday, June 22 nd 2009, 9.00h-12.00h. First read the entire examination. There

More information

IP and Mobility. Requirements to a Mobile IP. Terminology in Mobile IP

IP and Mobility. Requirements to a Mobile IP. Terminology in Mobile IP IP and Mobility Chapter 2 Technical Basics: Layer Methods for Medium Access: Layer 2 Chapter Wireless Networks: Bluetooth, WLAN, WirelessMAN, WirelessWAN Mobile Telecommunication Networks: GSM, GPRS, UMTS

More information

TRUFFLE Broadband Bonding Network Appliance. A Frequently Asked Question on. Link Bonding vs. Load Balancing

TRUFFLE Broadband Bonding Network Appliance. A Frequently Asked Question on. Link Bonding vs. Load Balancing TRUFFLE Broadband Bonding Network Appliance A Frequently Asked Question on Link Bonding vs. Load Balancing 5703 Oberlin Dr Suite 208 San Diego, CA 92121 P:888.842.1231 F: 858.452.1035 info@mushroomnetworks.com

More information

IERG 4080 Building Scalable Internet-based Services

IERG 4080 Building Scalable Internet-based Services Department of Information Engineering, CUHK Term 1, 2015/16 IERG 4080 Building Scalable Internet-based Services Lecture 4 Load Balancing Lecturer: Albert C. M. Au Yeung 30 th September, 2015 Web Server

More information

High Performance Cluster Support for NLB on Window

High Performance Cluster Support for NLB on Window High Performance Cluster Support for NLB on Window [1]Arvind Rathi, [2] Kirti, [3] Neelam [1]M.Tech Student, Department of CSE, GITM, Gurgaon Haryana (India) arvindrathi88@gmail.com [2]Asst. Professor,

More information

2.2 SIP-based Load Balancing. 3 SIP Load Balancing. 3.1 Proposed Load Balancing Solution. 2 Background Research. 2.1 HTTP-based Load Balancing

2.2 SIP-based Load Balancing. 3 SIP Load Balancing. 3.1 Proposed Load Balancing Solution. 2 Background Research. 2.1 HTTP-based Load Balancing SIP TRAFFIC LOAD BALANCING Ramy Farha School of Electrical and Computer Engineering University of Toronto Toronto, Ontario Email: rfarha@comm.utoronto.ca ABSTRACT This paper presents a novel solution to

More information

DESIGN AND DEVELOPMENT OF LOAD SHARING MULTIPATH ROUTING PROTCOL FOR MOBILE AD HOC NETWORKS

DESIGN AND DEVELOPMENT OF LOAD SHARING MULTIPATH ROUTING PROTCOL FOR MOBILE AD HOC NETWORKS DESIGN AND DEVELOPMENT OF LOAD SHARING MULTIPATH ROUTING PROTCOL FOR MOBILE AD HOC NETWORKS K.V. Narayanaswamy 1, C.H. Subbarao 2 1 Professor, Head Division of TLL, MSRUAS, Bangalore, INDIA, 2 Associate

More information

Application-layer protocols

Application-layer protocols Application layer Goals: Conceptual aspects of network application protocols Client server paradigm Service models Learn about protocols by examining popular application-level protocols HTTP DNS Application-layer

More information

Improving the Performance of HTTP over High Bandwidth-Delay Product circuits

Improving the Performance of HTTP over High Bandwidth-Delay Product circuits Improving the Performance of HTTP over High Bandwidth-Delay Product circuits A.J. McGregor National Laboratory for Applied Network Research San Diego Super Computer Center 10100 Hopkins Drive, San Diego,

More information

Agenda. Distributed System Structures. Why Distributed Systems? Motivation

Agenda. Distributed System Structures. Why Distributed Systems? Motivation Agenda Distributed System Structures CSCI 444/544 Operating Systems Fall 2008 Motivation Network structure Fundamental network services Sockets and ports Client/server model Remote Procedure Call (RPC)

More information

A Simulation Study of Effect of MPLS on Latency over a Wide Area Network (WAN)

A Simulation Study of Effect of MPLS on Latency over a Wide Area Network (WAN) A Simulation Study of Effect of MPLS on Latency over a Wide Area Network (WAN) Adeyinka A. Adewale, Samuel N. John, and Charles Ndujiuba 1 Department of Electrical and Information Engineering, Covenant

More information

Demand Routing in Network Layer for Load Balancing in Content Delivery Networks

Demand Routing in Network Layer for Load Balancing in Content Delivery Networks Demand Routing in Network Layer for Load Balancing in Content Delivery Networks # A SHRAVANI, 1 M.Tech, Computer Science Engineering E mail: sravaniathome@gmail.com # SYED ABDUL MOEED 2 Asst.Professor,

More information

Multicast vs. P2P for content distribution

Multicast vs. P2P for content distribution Multicast vs. P2P for content distribution Abstract Many different service architectures, ranging from centralized client-server to fully distributed are available in today s world for Content Distribution

More information

STANDPOINT FOR QUALITY-OF-SERVICE MEASUREMENT

STANDPOINT FOR QUALITY-OF-SERVICE MEASUREMENT STANDPOINT FOR QUALITY-OF-SERVICE MEASUREMENT 1. TIMING ACCURACY The accurate multi-point measurements require accurate synchronization of clocks of the measurement devices. If for example time stamps

More information

Internet Content Distribution

Internet Content Distribution Internet Content Distribution Chapter 4: Content Distribution Networks (TUD Student Use Only) Chapter Outline Basics of content distribution networks (CDN) Why CDN? How do they work? Client redirection

More information

DEPLOYMENT GUIDE Version 1.1. DNS Traffic Management using the BIG-IP Local Traffic Manager

DEPLOYMENT GUIDE Version 1.1. DNS Traffic Management using the BIG-IP Local Traffic Manager DEPLOYMENT GUIDE Version 1.1 DNS Traffic Management using the BIG-IP Local Traffic Manager Table of Contents Table of Contents Introducing DNS server traffic management with the BIG-IP LTM Prerequisites

More information

Chapter 4. VoIP Metric based Traffic Engineering to Support the Service Quality over the Internet (Inter-domain IP network)

Chapter 4. VoIP Metric based Traffic Engineering to Support the Service Quality over the Internet (Inter-domain IP network) Chapter 4 VoIP Metric based Traffic Engineering to Support the Service Quality over the Internet (Inter-domain IP network) 4.1 Introduction Traffic Engineering can be defined as a task of mapping traffic

More information

IP SLAs Overview. Finding Feature Information. Information About IP SLAs. IP SLAs Technology Overview

IP SLAs Overview. Finding Feature Information. Information About IP SLAs. IP SLAs Technology Overview This module describes IP Service Level Agreements (SLAs). IP SLAs allows Cisco customers to analyze IP service levels for IP applications and services, to increase productivity, to lower operational costs,

More information

IP Networking. Overview. Networks Impact Daily Life. IP Networking - Part 1. How Networks Impact Daily Life. How Networks Impact Daily Life

IP Networking. Overview. Networks Impact Daily Life. IP Networking - Part 1. How Networks Impact Daily Life. How Networks Impact Daily Life Overview Dipl.-Ing. Peter Schrotter Institute of Communication Networks and Satellite Communications Graz University of Technology, Austria Fundamentals of Communicating over the Network Application Layer

More information

VoIP versus VoMPLS Performance Evaluation

VoIP versus VoMPLS Performance Evaluation www.ijcsi.org 194 VoIP versus VoMPLS Performance Evaluation M. Abdel-Azim 1, M.M.Awad 2 and H.A.Sakr 3 1 ' ECE Department, Mansoura University, Mansoura, Egypt 2 ' SCADA and Telecom General Manager, GASCO,

More information

Alteon Global Server Load Balancing

Alteon Global Server Load Balancing Alteon Global Server Load Balancing Whitepaper GSLB Operation Overview Major Components Distributed Site Monitoring Distributed Site State Protocol Internet Topology Awareness DNS Authoritative Name Server

More information

Investigation and Comparison of MPLS QoS Solution and Differentiated Services QoS Solutions

Investigation and Comparison of MPLS QoS Solution and Differentiated Services QoS Solutions Investigation and Comparison of MPLS QoS Solution and Differentiated Services QoS Solutions Steve Gennaoui, Jianhua Yin, Samuel Swinton, and * Vasil Hnatyshin Department of Computer Science Rowan University

More information

Course Overview: Learn the essential skills needed to set up, configure, support, and troubleshoot your TCP/IP-based network.

Course Overview: Learn the essential skills needed to set up, configure, support, and troubleshoot your TCP/IP-based network. Course Name: TCP/IP Networking Course Overview: Learn the essential skills needed to set up, configure, support, and troubleshoot your TCP/IP-based network. TCP/IP is the globally accepted group of protocols

More information

Secure SCTP against DoS Attacks in Wireless Internet

Secure SCTP against DoS Attacks in Wireless Internet Secure SCTP against DoS Attacks in Wireless Internet Inwhee Joe College of Information and Communications Hanyang University Seoul, Korea iwjoe@hanyang.ac.kr Abstract. The Stream Control Transport Protocol

More information

Lecture Computer Networks

Lecture Computer Networks Prof. Dr. H. P. Großmann mit M. Rabel sowie H. Hutschenreiter und T. Nau Sommersemester 2012 Institut für Organisation und Management von Informationssystemen Thomas Nau, kiz Lecture Computer Networks

More information

IPv4 and IPv6 Integration. Formation IPv6 Workshop Location, Date

IPv4 and IPv6 Integration. Formation IPv6 Workshop Location, Date IPv4 and IPv6 Integration Formation IPv6 Workshop Location, Date Agenda Introduction Approaches to deploying IPv6 Standalone (IPv6-only) or alongside IPv4 Phased deployment plans Considerations for IPv4

More information

Advanced Networking Technologies

Advanced Networking Technologies Advanced Networking Technologies Chapter 14 Navigating Content Networks (Acknowledgement: These slides have been prepared by Dr.-Ing. Markus Hofmann) Advanced Networking (SS 15): 14 Navigating Content

More information

Requirements of Voice in an IP Internetwork

Requirements of Voice in an IP Internetwork Requirements of Voice in an IP Internetwork Real-Time Voice in a Best-Effort IP Internetwork This topic lists problems associated with implementation of real-time voice traffic in a best-effort IP internetwork.

More information

Guide to DDoS Attacks December 2014 Authored by: Lee Myers, SOC Analyst

Guide to DDoS Attacks December 2014 Authored by: Lee Myers, SOC Analyst INTEGRATED INTELLIGENCE CENTER Technical White Paper William F. Pelgrin, CIS President and CEO Guide to DDoS Attacks December 2014 Authored by: Lee Myers, SOC Analyst This Center for Internet Security

More information

TCP over Multi-hop Wireless Networks * Overview of Transmission Control Protocol / Internet Protocol (TCP/IP) Internet Protocol (IP)

TCP over Multi-hop Wireless Networks * Overview of Transmission Control Protocol / Internet Protocol (TCP/IP) Internet Protocol (IP) TCP over Multi-hop Wireless Networks * Overview of Transmission Control Protocol / Internet Protocol (TCP/IP) *Slides adapted from a talk given by Nitin Vaidya. Wireless Computing and Network Systems Page

More information