Part one of the e-Book, The Ultimate Guide to Gaining Control of the WAN, explains the drivers leading to the wide...
By submitting your email address, you agree to receive emails regarding relevant topic offers from TechTarget and its partners. You can withdraw your consent at any time. Contact TechTarget at 275 Grove Street, Newton, MA.
area network's (WAN’s) critical role in lowering IT costs and examines how branch office server consolidation can help enterprises manage traffic, increase performance and get WAN costs under control, whether you do it yourself or use a service provider. Navigate the table of contents to read other sections of this e-Book or skip below to learn more about achieving WAN cost savings through branch office server consolidation.
Table of contents
Part 1: Save WAN costs with branch office server consolidation
Part 2: How to accelerate encrypted traffic using WAN optimization
Part 3: Virtual desktop infrastructure problems solved by WAN optimization
Part 4: Using the WAN for telepresence, video conferencing
Part 5: Using WAN optimization for bandwidth management and monitoring
Part 6: Update network security architecture during server consolidation
Part 7: Wide area network optimization: Do it in-house, or use a WAN service provider?
Understanding branch office server consolidation
The most common and biggest money-saving project that enterprises are embarking on is server consolidation. Moving branch office servers to a central data center saves money in many ways. The first is by reducing hardware cost. Branch office server utilization is generally below 25%, and there is little that can be done to improve that. Moving several servers to the data center and using virtualization software allows multiple branch office servers to be consolidated into one physical server. Server consolidation saves on hardware cost and software licenses and provides a greener and lower-cost solution because it requires less power and cooling. It also makes maintenance easier because all the servers are located in one place. Disaster recovery becomes easier, as well, because the servers easily fit into the data center’s disaster-recovery scheme, rather than being dispersed. Overall, consolidating branch office servers in the data center is a winner and a trend that will become even more common in a down economy.
The problem with a server consolidation project is its effect on the WAN. Data that was local to the user and retrieved over a high-speed local area network (LAN) now must cross the WAN. Normally, this means that bandwidth must be increased. But even with increased bandwidth, there can be response-time problems that speed alone will not solve. The main problem is the Common Internet File System (CIFS), the protocol Microsoft uses to allow programs to make requests for files and services on remote computers. Initially, it was very inefficient over the WAN. For example, when CIFS downloaded a file, it asked for a block of data and then waited for an acknowledgment before asking for the next block of data. This inefficient protocol was not a problem over a LAN because it involved almost no latency. Over the WAN, it created a slow start-stop process that greatly increased file download time. Recent improvements to CIFS allow it to ask for multiple blocks of data at once, increasing its efficiency. That has not completely eliminated the start-stop process over the WAN and long downloads, however.
The problem with a server consolidation project is its effect on the WAN. Data that was local on a high-speed LAN must now cross the WAN.Another problem that server consolidation introduces is the effect on some of the services that the servers provide to local users.
Local servers provide Dynamic Host Configuration Protocol (DHCP), domain name system (DNS), and activity directory services and support printing to local printers. Moving these services to the data center does not enhance them or provide any significant benefit and instead can introduce problems. If the link to the data center is down, then branch office users are prevented from performing any networking functions. Even if there are still local servers or alternate routes to the Internet, users can’t take advantage of them without DHCP or DNS. Having a good backup path in the WAN is an important strategy, and if problems do occur it will be necessary to have DHCP and DNS functions locally. Even if the link is up, routing a print job to the data center and then back to the branch office’s local printer is a waste of resources.
However, these problems and their impact on the WAN or local services are not reason enough to stop server consolidation when the total cost savings picture is considered. Instead, WAN managers must find ways around the problems.
Overcoming server consolidation challenges
The answer is acceleration and optimization. This is accomplished with devices commonly referred to as WAN optimization controllers (WOCs).WAN optimization is a symmetrical solution, which means a WOC must be installed in each branch office and in the data center. The WOC in the data center supports multiple branch office WOCs, with the exact ratio of branch office devices to data center devices depending on the capacity of the data center WOC.
WAN bandwidth optimization tools solve data center problems with a combination of technologies. The need for increased bandwidth is solved by compressing data using a technique commonly referred to as dictionary compression. Dictionary compression can easily reduce the amount of data sent by a factor of five to 10 times, with reduction rates as high as 20 to 30 times. Lines running at 75% capacity can be reduced to less than 10%. Server consolidation bandwidth requirements can fit within the existing bandwidth. This is not hype; independent industry tests and user experiences have confirmed that these rates and reductions are possible.
Dictionary compression achieves its high reduction rates by learning patterns in the data and substituting a reference number for the patterns. As data is passing through the WOC, it breaks the data into a series of patterns and stores them, creating a dictionary of patterns. When the next message comes through, the WOC checks to see whether any of its stored patterns matches patterns in the message. If it finds them, the WOC removes the patterns and substitutes reference numbers in their place. The WOC on the receiving end removes the reference number and replaces it with its copy of the pattern from its dictionary. The patterns do not have to be exchanged between the two WOCs because they are learning and storing the exact same pattern, using the same reference number independently.
WAN managers must carefully understand how appliances handle SSL traffic to ensure it is the best fit for their environment.
For example, if a user retrieves a file from a server that he accessed the day before, then the WOC need only send a series of reference numbers. That’s because the file was turned into a pattern the first time it was retrieved. If a section of the file has changed since it was last retrieved, then the WOC sends the reference numbers for the part that hasn’t changed, along with the small part that has. This example uses the same file but the patterns can come from other messages as well; the WOC uses its entire dictionary of patterns to find a match. Replacing patterns with reference numbers also reduces the number of packets sent over the network because the WOC can bundle reference numbers from multiple packets into one.
Dictionary compression works only on uncompressed data or un-encrypted data. There is no reduction with video and voice since those media are already compressed. The best action for WAN optimization is not to waste time trying to compress it.
Continue reading part two of this e-Book to learn how to accelerate encrypted traffic using WAN optimization during a server consolidation project.
About the author:
Robin Layland is President of Layland Consulting. As an industry analyst and consultant, Robin has covered all aspects of networking from both the business and technical side, and has published over 100 articles in leading trade journals including NetworkWorld, Business Communication Review, Network Magazine and Data Communications. Prior to his current role, Robin spent a combined fifteen years at American Express and Travelers Insurance in a wide range of jobs including network architect, technical support, management, programming, performance analysis and capacity planning.