WAN Acceleration Technology - WAN Optimization
Bandwidth and Latency are two of the major problems in Wide Area Network (WAN) communications. Information Technology (IT) based Enterprises have been confronted with these challenges in the face of remote office network connectivity. A lot of emphasis is given for connectivity since it is crucial for network communications, application access, security, backup, messaging, etc. While bandwidth refers to the magnitude of the connection, latency is the time taken for a data packet to traverse from one point in the network to another.
Enterprises are requiring more bandwidth with the increasing volume of business-critical data transmission over the network. Latency is found in many networks due to voluminous data, making the transit time longer. As a result, many companies increase the cost of systems and develop a distributed environment in order to reduce cost of communications. Many companies maintain multiple sites and similar systems at each location since it is more economical than increasing the network link size between sites. Optimization of WAN bandwidth is one of the popular and recognized methods used by organizations to have optimized networks as well as save costs. This is possible by using a WAN Accelerator, which is a key component in WAN bandwidth optimization.
WAN Accelerators have come in at the right time to address the networking needs of businesses. Being a new generation of communication hardware devices, WAN Accelerators are installed at both ends of a communication link. They can thereby manage all Internet Protocol (IP) traffic that is transferred across those links. A typical function of WAN accelerators is improving bandwidth usage and accelerating network performance. They can also be tuned to perform other specific applications. Businesses are seen mushrooming across a wider geography, and hence necessitate better connectivity for uninterrupted business operations. Lower latency and higher throughput enable WAN optimization acceleration and render productive results.
WAN Accelerator appliances, besides being used for general acceleration, are also used for application acceleration. They are capable of performing the following functions:
- Data compression
- Compression of IP headers
- Acceleration of Transmission Control Protocol (TCP)
- Acceleration of CIFS (Common Internet File System)
- Mitigation of lost data packets
- Forward Error Correction (FEC)
- Caching
- Enforcing Quality of Service (QoS)
WAN Accelerators use a few key technologies such as compression, direct caching, pattern matching, etc. to better the network performance. In Compression method, these devices eliminate surplus unnecessary data from the communication stream, thereby reducing the size of data packets that are being transmitted and received. Direct Caching is a technique used by a WAN Accelerator where replication of the data traffic passing over a WAN circuit is taken. The data that has already passed through the circuit will not be sent again. The transmitting WAN Accelerator informs the receiving WAN Accelerator about the data and then the receiving WAN Accelerator retrieves the information from the cached memory. This working is similar to that of an Internet Proxy server.
As technology takes a new dimension, these techniques are further improvised. An enhanced technique than direct caching emerged – Pattern Matching. Certain technology vendors have devised methods to categorize data passing through a WAN Accelerator. The data instead of being transmitted as such is substituted for a key based on that particular pattern recognizable by the receiving WAN Accelerator. It is then aligned in the correct data pattern at the receiving link. This task is carried out in small amounts of data. Advanced WAN Accelerators enable WAN Optimization acceleration through inserting data packets into the communication stream at the far side of the link for the client to view. This substantially increases speed. By such a technique as this, a WAN accelerator might be able to insert 1MB of data into a 20 byte stream, and accelerate performance.
