WAN optimization

Last updated

WAN optimization is a collection of techniques for improving data transfer across wide area networks (WANs). In 2008, the WAN optimization market was estimated to be $1 billion, [1] and was to grow to $4.4 billion by 2014 according to Gartner, [2] a technology research firm. In 2015 Gartner estimated the WAN optimization market to be a $1.1 billion market. [3]

The most common measures of TCP data-transfer efficiencies (i.e., optimization) are throughput, bandwidth requirements, latency, protocol optimization, and congestion, as manifested in dropped packets. [4] In addition, the WAN itself can be classified with regards to the distance between endpoints and the amounts of data transferred. Two common business WAN topologies are Branch to Headquarters and Data Center to Data Center (DC2DC). In general, "Branch" WAN links are closer, use less bandwidth, support more simultaneous connections, support smaller connections and more short-lived connections, and handle a greater variety of protocols. They are used for business applications such as email, content management systems, database application, and Web delivery. In comparison, "DC2DC" WAN links tend to require more bandwidth, are more distant, and involve fewer connections, but those connections are bigger (100 Mbit/s to 1 Gbit/s flows) and of longer duration. Traffic on a "DC2DC" WAN may include replication, back up, data migration, virtualization, and other Business Continuity/Disaster Recovery (BC/DR) flows.

WAN optimization has been the subject of extensive academic research almost since the advent of the WAN. [5] In the early 2000s, research in both the private and public sectors turned to improving the end-to-end throughput of TCP, [6] and the target of the first proprietary WAN optimization solutions was the Branch WAN. In recent years, however, the rapid growth of digital data, and the concomitant needs to store and protect it, has presented a need for DC2DC WAN optimization. For example, such optimizations can be performed to increase overall network capacity utilization, [7] [8] meet inter-datacenter transfer deadlines, [9] [10] [11] or minimize average completion times of data transfers. [11] [12] As another example, private inter-datacenter WANs can benefit optimizations for fast and efficient geo-replication of data and content, such as newly computed machine learning models or multimedia content. [13] [14]

Component techniques of Branch WAN Optimization include deduplication, wide area file services (WAFS), SMB proxy, HTTPS Proxy, media multicasting, web caching, and bandwidth management. Requirements for DC2DC WAN Optimization also center around deduplication and TCP acceleration, however these must occur in the context of multi-gigabit data transfer rates.

WAN optimization techniques

Deduplication
Eliminates the transfer of redundant data across the WAN by sending references instead of the actual data. By working at the byte level, benefits are achieved across IP applications.
Data compression
Relies on data patterns that can be represented more efficiently. Essentially compression techniques similar to ZIP, RAR, ARJ etc. are applied on-the-fly to data passing through hardware (or virtual machine) based WAN acceleration appliances.
Latency optimization
Can include TCP refinements such as window-size scaling, selective acknowledgements, Layer 3 congestion control algorithms, and even co-location strategies in which the application is placed in near proximity to the endpoint to reduce latency. [15] In some implementations, the local WAN optimizer will answer the requests of the client locally instead of forwarding the request to the remote server in order to leverage write-behind and read-ahead mechanisms to reduce WAN latency.
Caching/proxy
Staging data in local caches; Relies on human behavior, accessing the same data over and over.
Forward error correction
Mitigates packet loss by adding another loss-recovery packet for every N packets that are sent, and this would reduce the need for retransmissions in error-prone and congested WAN links.
Protocol spoofing
Bundles multiple requests from chatty applications into one. May also include stream-lining protocols such as CIFS.
Traffic shaping
Controls data flow for specific applications. Giving flexibility to network operators/network admins to decide which applications take precedence over the WAN. A common use case of traffic shaping would be to prevent one protocol or application from hogging or flooding a link over other protocols deemed more important by the business/administrator. Some WAN acceleration devices are able to traffic shape with granularity far beyond traditional network devices. Such as shaping traffic on a per-user and per-application basis simultaneously.
Equalizing
Makes assumptions on what needs immediate priority based on the data usage. Usage examples for equalizing may include wide open unregulated Internet connections and clogged VPN tunnels.
Connection limits
Prevents access gridlock in and to denial of service or to peer. Best suited for wide open Internet access links, can also be used links.
Simple rate limits
Prevents one user from getting more than a fixed amount of bandwidth. Best suited as a stop gap first effort for remediating a congested Internet connection or WAN link.

Related Research Articles

Routing is the process of selecting a path for traffic in a network or between or across multiple networks. Broadly, routing is performed in many types of networks, including circuit-switched networks, such as the public switched telephone network (PSTN), and computer networks, such as the Internet.

The Transmission Control Protocol (TCP) is one of the main protocols of the Internet protocol suite. It originated in the initial network implementation in which it complemented the Internet Protocol (IP). Therefore, the entire suite is commonly referred to as TCP/IP. TCP provides reliable, ordered, and error-checked delivery of a stream of octets (bytes) between applications running on hosts communicating via an IP network. Major internet applications such as the World Wide Web, email, remote administration, and file transfer rely on TCP, which is part of the Transport layer of the TCP/IP suite. SSL/TLS often runs on top of TCP.

<span class="mw-page-title-main">Frame Relay</span> Wide area network technology

Frame Relay is a standardized wide area network (WAN) technology that specifies the physical and data link layers of digital telecommunications channels using a packet switching methodology. Originally designed for transport across Integrated Services Digital Network (ISDN) infrastructure, it may be used today in the context of many other network interfaces.

<span class="mw-page-title-main">Telecommunications network</span> Network for communications over distance

A telecommunications network is a group of nodes interconnected by telecommunications links that are used to exchange messages between the nodes. The links may use a variety of technologies based on the methodologies of circuit switching, message switching, or packet switching, to pass messages and signals.

Traffic shaping is a bandwidth management technique used on computer networks which delays some or all datagrams to bring them into compliance with a desired traffic profile. Traffic shaping is used to optimize or guarantee performance, improve latency, or increase usable bandwidth for some kinds of packets by delaying other kinds. It is often confused with traffic policing, the distinct but related practice of packet dropping and packet marking.

Network congestion in data networking and queueing theory is the reduced quality of service that occurs when a network node or link is carrying more data than it can handle. Typical effects include queueing delay, packet loss or the blocking of new connections. A consequence of congestion is that an incremental increase in offered load leads either only to a small increase or even a decrease in network throughput.

An overlay network is a computer network that is layered on top of another network. The concept of overlay networking is distinct from the traditional model of OSI layered networks, and almost always assumes that the underlay network is an IP network of some kind.

Network performance refers to measures of service quality of a network as seen by the customer.

Transmission Control Protocol (TCP) uses a congestion control algorithm that includes various aspects of an additive increase/multiplicative decrease (AIMD) scheme, along with other schemes including slow start and congestion window (CWND), to achieve congestion avoidance. The TCP congestion-avoidance algorithm is the primary basis for congestion control in the Internet. Per the end-to-end principle, congestion control is largely a function of internet hosts, not the network itself. There are several variations and versions of the algorithm implemented in protocol stacks of operating systems of computers that connect to the Internet.

TCP tuning techniques adjust the network congestion avoidance parameters of Transmission Control Protocol (TCP) connections over high-bandwidth, high-latency networks. Well-tuned networks can perform up to 10 times faster in some cases. However, blindly following instructions without understanding their real consequences can hurt performance as well.

Packet loss occurs when one or more packets of data travelling across a computer network fail to reach their destination. Packet loss is either caused by errors in data transmission, typically across wireless networks, or network congestion. Packet loss is measured as a percentage of packets lost with respect to packets sent.

A middlebox is a computer networking device that transforms, inspects, filters, and manipulates traffic for purposes other than packet forwarding. Examples of middleboxes include firewalls, network address translators (NATs), load balancers, and deep packet inspection (DPI) devices.

Packeteer, Inc., founded in 1996 by Robert Packer, Brett Galloway and Bob Luxenberg was an I.T. company based in Cupertino, California that was listed on the NASDAQ. Networking appliances focus on Application Traffic Management and optimization for wide area networks. They held at least 40 patents for various network optimization methods. Packeteer was acquired by Blue Coat Systems in 2008.

An application delivery network (ADN) is a suite of technologies that, when deployed together, provide availability, security, visibility, and acceleration for Internet applications such as websites. ADN components provide supporting functionality that enables website content to be delivered to visitors and other users of that website, in a fast, secure, and reliable way.

A reliable multicast is any computer networking protocol that provides a reliable sequence of packets to multiple recipients simultaneously, making it suitable for applications such as multi-receiver file transfer.

Zeta-TCP refers to a set of proprietary Transmission Control Protocol (TCP) algorithms aiming at improving the end-to-end performance of TCP, regardless of whether the peer is Zeta-TCP or any other TCP protocol stack, in other words, to be compatible with the existing TCP algorithms. It was designed and implemented by AppEx Networks Corporation.

Infineta Systems was a company that made WAN optimization products for high performance, latency-sensitive network applications. The company advertised that it allowed application data rate to exceed the nominal data rate of the link. Infineta Systems ceased operations by February 2013, a liquidator was appointed, and its products will no longer be manufactured, sold or distributed.

Software-defined networking (SDN) is an approach to network management that enables dynamic and programmatically efficient network configuration to improve network performance and monitoring in a manner more akin to cloud computing than to traditional network management. SDN is meant to improve the static architecture of traditional networks and may be employed to centralize network intelligence in one network component by disassociating the forwarding process of network packets from the routing process. The control plane consists of one or more controllers, which are considered the brains of the SDN network, where the whole intelligence is incorporated. However, centralization has certain drawbacks related to security, scalability and elasticity.

Admission control is a validation process in communication systems where a check is performed before a connection is established to see if current resources are sufficient for the proposed connection.

A software-defined wide area network (SD-WAN) is a wide area network that uses software-defined networking technology, such as communicating over the Internet using overlay tunnels which are encrypted when destined for internal organization locations.

References

  1. Machowinski, Matthias. "WAN optimization market passes $1 billion in 2008, up 29%; enterprise router market down". Enterprise Routers and WAN Optimization Appliances. Infonetics Research. Retrieved 19 July 2011.
  2. Skorupa, Joe; Severine Real (2010). "Forecast: Application Acceleration Equipment, Worldwide, 2006–2014, 2Q10 Update". Gartner, Inc. Retrieved 19 July 2011.[ dead link ]
  3. Munch, Bjarne; Neil Rickard (2015). "Magic Quadrant for WAN Optimization, 17 March 2015". Gartner, Inc. Retrieved 26 March 2015.
  4. Cardwell, N.; Savage, S.; Anderson, T. (2000). "Modeling TCP latency". Proceedings IEEE INFOCOM 2000. Conference on Computer Communications. Nineteenth Annual Joint Conference of the IEEE Computer and Communications Societies (Cat. No.00CH37064). Vol. 3. Dept. of Comput. Sci. & Eng., Washington Univ., Seattle, WA: IEEE.org. pp. 1742–1751. doi:10.1109/INFCOM.2000.832574. ISBN   0-7803-5880-5. S2CID   6581992.
  5. Jacobson, Van (October 1988). "TCP Extensions for Long-Delay Paths". Request for Comments: 1072. Internet Engineering Task Force (IETF). Retrieved 19 July 2011.
  6. Floyd, Sally. "HighSpeed TCP for Large Congestion Windows". Request for Comments: 3649. Internet Engineering Task Force (IETF). Retrieved 19 July 2011.
  7. S. Jain; et al. (2013). "B4: Experience with a Globally-Deployed Software Defined WAN" (PDF). Retrieved April 4, 2018.
  8. C. Hong; et al. (2013). "Achieving High Utilization with Software-Driven WAN". Microsoft . Retrieved April 4, 2018.
  9. S. Kandula; et al. (2014). "Calendaring for Wide Area Networks" (PDF). Microsoft . Retrieved April 4, 2018.
  10. M. Noormohammadpour; et al. (2016). "DCRoute: Speeding up Inter-Datacenter Traffic Allocation while Guaranteeing Deadlines" . Retrieved April 4, 2018.
  11. 1 2 X. Jin; et al. (2016). "Optimizing Bulk Transfers with Software-Defined Optical WAN" (PDF). Retrieved April 4, 2018.
  12. M. Noormohammadpour; et al. (2018). "Minimizing Flow Completion Times using Adaptive Routing over Inter-Datacenter Wide Area Networks" . Retrieved April 4, 2018.
  13. M. Noormohammadpour; et al. (July 10, 2017). "DCCast: Efficient Point to Multipoint Transfers Across Datacenters". USENIX. Retrieved July 26, 2017.
  14. M. Noormohammadpour; et al. (2018). "QuickCast: Fast and Efficient Inter-Datacenter Transfers using Forwarding Tree Cohorts" . Retrieved January 23, 2018.
  15. Paris, Chandler. "Latency & Colocation" . Retrieved 20 July 2011.