Contents

Configuring WAAS·· 1

About WAAS· 1

TFO·· 1

DRE· 2

LZ compression· 2

Protocols and standards· 3

Restrictions and guidelines: WAAS configuration· 3

Configuring a WAAS instance· 3

Directing traffic to a WAAS instance through a QoS policy· 4

Display and maintenance commands for WAAS· 5

 

Configuring WAAS

About WAAS

The Wide Area Application Services (WAAS) feature is a set of services that can optimize WAN traffic. WAAS solves WAN issues such as high delay and low bandwidth by using optimization services.

TFO

TFO optimizes TCP traffic without modifying packet header information. TFO uses the following optimization methods:

·     Slow start optimization.

·     Increased buffering.

·     Congestion algorithm optimization.

·     Selective acknowledgement.

Slow start optimization

The initial congestion window size for TCP slow start is one TCP segment. During slow start, TCP doubles the congestion window size for each received ACK that acknowledges new data. In this manner, the congestion window will reach an appropriate value by examining the congestion status. In a WAN environment, the congestion window takes a long time to reach an appropriate size because of high delay.

Slow start optimization shortens the slow start process by increasing the initial congestion window size.

Increased buffering

TCP has a maximum buffer size of 64 KB. After the sender sends 64 KB data, it must wait for an ACK from the receiver before continuing to send data. This mechanism wastes bandwidth on a WAN link.

Increased buffering increases the TCP buffer size to a maximum of 16384 KB. This improves link efficiency.

Congestion algorithm optimization

TCP uses the congestion window to control congestion. The window size indicates the size of data that can be sent out before an ACK is received. The window size changes with the congestion status. The greater the window size, the faster the data rate. A higher data rate more likely causes congestion. The smaller the window size, the lower the data rate. A lower data rate causes low link efficiency.

Congestion algorithm optimization achieves a trade-off between the data rate and congestion by selecting the optimum window size.

Selective acknowledgement

TCP uses a cumulative acknowledgement scheme. This scheme forces the sender to either wait a roundtrip time to know each lost packet, or to unnecessarily retransmit segments that have been correctly received. When multiple nonconsecutive segments are lost, this scheme reduces overall TCP throughput.

Selective acknowledgement (SACK) allows the receiver to inform the sender of all segments that have arrived successfully. The sender retransmits only the segments that have been lost.

DRE

DRE reduces the size of transmitted data by replacing repeated data blocks with shorter indexes. A WAAS device synchronizes its data dictionary to its peer devices. A data dictionary stores mappings between repeated data blocks and indexes.

Replacing repeated data blocks with indexes is called DRE compression. Replacing indexes with repeated data blocks is called DRE decompression.

DRE compression process

DRE compresses data in the following process:

1.     The sending WAAS device caches TCP data and sends a large data block to the DRE module.

2.     The DRE module divides the large data block into non-overlapping data blocks.

¡     For a repeated data block, the DRE module performs the following operations:

-     Replaces the data block with its index and creates an MD5 digest for the data block.

-     Sends the index and MD5 digest to the peer.

¡     For a non-repeated data block, the DRE module performs the following operations:

-     Creates an index for the data block and adds it to the local data dictionary.

-     Creates an MD5 digest for the data block and sends the data block, index, and MD5 digest to the peer.

To improve calculation speed and efficiency, DRE uses a sliding window mechanism to segment data and detect data redundancy. This mechanism detects repeated data blocks by using a fixed-size window to compare the original data byte by byte with data blocks in the dictionary.

DRE decompression process

DRE decompresses data in the following process:

1.     The receiving WAAS device reconstructs the original data.

¡     For an index, the device replaces the index with its data block after querying the data dictionary.

If the query fails, the decompression fails, and the receiving WAAS device waits for the peer to retransmit the data.

¡     For an index and a data block, the device creates an entry for them and adds the entry to the local data dictionary.

2.     The receiving WAAS device calculates an MD5 digest for the original data and compares the calculated MD5 digest with the MD5 digest in the packet.

¡     If the two MD5 digests are the same, the decompression succeeds.

¡     If the two MD5 digests are different, the decompression fails, and the receiving WAAS device waits for the peer to retransmit the data.

LZ compression

LZ compression is a lossless compression algorithm that uses a compression dictionary to replace repeated data in the same message. The compression dictionary is carried in the compression result. The sending device uses the sliding window technology to detect repeated data.

Compared with DRE, LZ compression has a lower compression ratio. LZ compression does not require synchronization of compression dictionaries between the local and peer devices. This reduces memory consumption.

Protocols and standards

·     RFC 1323, TCP Extensions for High Performance

·     RFC 3390, Increasing TCP's Initial Window

·     RFC 2581, TCP Congestion Control

·     RFC 2018, TCP Selective Acknowledgment Options

·     RFC 3042, Enhancing TCP's Loss Recovery Using Limited Transmit

·     RFC 2582, The NewReno Modification to TCP's Fast Recovery Algorithm

Restrictions and guidelines: WAAS configuration

For the WAAS feature to work correctly, make sure fast forwarding load sharing is disabled. For information about fast forwarding load sharing, see Layer 3—IP Services Configuration Guide.

Configuring a WAAS instance

About this task

After you associate a WAAS instance with a service instance group and configure an optimization action, the device directs matching traffic to the primary node or a secondary node in the service instance group for optimization. A WAAS instance is used in a QoS policy. The QoS policy identifies the traffic to be optimized. The optimization action configured in the WAAS instance is taken on the traffic.

Restrictions and guidelines

You can associate a nonexistent service instance group. However, for the configuration to take effect, you must create the service instance group by using the service-instance-group command. For more information about service instance groups, see High Availability Configuration Guide.

Procedure

1.     Enter system view.

system-view

2.     Create a WAAS instance and enter WAAS instance view.

waas instance instance-name [ id id ]

3.     Associate the WAAS instance with a service instance group.

service-instance-group group-name

By default, a WAAS instance is not associated with any service instance group.

4.     Configure an optimization action.

optimize { compress | decompress }

By default, no optimization action is configured.

5.     Return to system view.

quit

Directing traffic to a WAAS instance through a QoS policy

About this task

You can use a QoS policy on a border device in a WAN to directing specific traffic to a WAAS instance for traffic optimization. The WAAS instance compresses outgoing traffic and decompresses incoming traffic.

Restrictions and guidelines

You can apply a QoS policy containing a WAAS instance binding action to an interface or globally. The QoS policy applied globally has a higher priority.

Procedure

1.     Enter system view.

system-view

2.     Create a traffic class and enter traffic class view.

traffic classifier classifier-name [ operator { and | or } ]

3.     Configure a match criterion.

if-match [ not ] match-criteria

By default, no match criterion is configured.

For more information about match criteria, see the if-match command in ACL and QoS Command Reference.

4.     Return to system view.

quit

5.     Create a traffic behavior and enter traffic behavior view.

traffic behavior behavior-name

6.     Bind a WAAS instance.

bind waas-instance instance-name

7.     Return to system view.

quit

8.     Configure the high-priority QoS policy.

For detailed steps to configure a QoS policy, see QoS commands in ACL and QoS Command Reference.

9.     Return to system view.

quit

10.     Create a QoS policy and enter QoS policy view.

qos policy policy-name

11.     Associate a traffic class with a traffic behavior to create a class-behavior association in the QoS policy.

classifier classifier-name behavior behavior-name

For more information, see QoS commands in ACL and QoS Command Reference.

12.     Return to system view.

quit

13.     Apply the QoS policy.

¡     Apply the QoS policy to an interface.

interface interface-type interface-number

qos apply policy policy-name { inbound | outbound }

¡     Apply the QoS policy globally.

qos apply policy policy-name { inbound | outbound }

For more information applying a QoS policy, see QoS configuration in ACL and QoS Configuration Guide.

Display and maintenance commands for WAAS

Execute display commands in any view and reset commands in user view.

 

Task	Command
Display compression statistics.	display waas compress statistics [ slot slot-number ]
Display WAAS instance configuration.	display waas instance [ instance-name instance-name ]
Clear compression statistics.	reset waas compress statistics [ slot slot-number ]