Technological breakthroughs in WAN Optimization
In my last blog post, I talked about the state of the WAN Optimization market and the issues faced by the incumbent vendors. Let’s now look at the different ways in which these vendors try to solve such problems, and how Aryaka approaches the issues a little differently and a lot more effectively! In the context of an enterprise WAN, most large enterprises have a hub-spoke architecture (with one or more hubs) to manage things like access, data management, compliance and security.
TCP Optimization: TCP Optimization solves issues with the network due to limitations of the TCP protocol as well as the long distances across which data must be transferred. In the appliance model, these issues are addressed across a single link that consists of links with varying characteristics. The appliance takes into account the worst case scenario across these links and optimizes accordingly. Aryaka has a unique way of achieving the highest levels of TCP optimization possible across an enterprise network. Each link is divided into three segments- HQ to the Aryaka POP, Aryaka POP to POP and Aryaka POP to branch. Each leg is optimized with appropriate settings and techniques. The strategy to optimize a high-bandwidth low-latency last-mile link will be different from that required to optimize a low-bandwidth high-latency intercontinental core link.
Compression: Compression is really needed on the low speed spoke links, where even a single file will take a long time to be transferred over the network and choke the link at peak loads. At the hub, site-to-site traffic is typically a small part of the total link bandwidth because of extra capacity needed for hosted data connections, internet access and more. Bandwidth at hub locations is relatively more affordable as hubs are in larger cities and more developed geographies. Aryaka also has an optional fully managed appliance for low-bandwidth high-latency links in remote geographies to support compression.
A second aspect of compression at Aryaka within our POPs is accomplished through our proprietary patent-pending ARRTM. Advanced Redundancy Removal (ARRTM) reduces the amount of data transferred across links using pattern matching on the data and ensuring data which is recognized is not re-transmitted, thereby saving valuable bandwidth, especially in remote geographies and on high-latency low-bandwidth links.
Application Proxies: Application-level proxies are used to solve application proxy issues caused by usage of ‘chatty protocols’ and application latency. These proxies are also the biggest headache for users in terms of configuration, tuning, management, upgrades, OS and application version-specific differences and more. To resolve this headache, all proxies are optimized within our distributed POP infrastructure, where we can easily manage and configure them for maximum performance. Any issues can be resolved with real time response as all POPs are monitored from our dedicated Network Operations Center (NOC). Traditional appliances on the other hand get outdated, need regular software/maintenance upgrades and have limitations across the board. The Aryaka platform is built to be the future of WAN optimization.
Next up: The importance of the visibility piece to the WAN Optimization puzzle.