Automation in SD-WAN and why you need WAN acceleration

Robert Sturt, general manager of streaming service provider Netify, pointed out that software-defined wide area network (SD-WAN) is a great technology. He believes that this is the most fundamental change in networking since MPLS replaced frame relay and asynchronous transfer mode. He believes that SD-WAN is the beginning of the journey of the technology stack, and their integration with artificial intelligence and automation represents "the next stage of leveraging the extensive insights and data that SD-WAN systems have already generated." One example is the use of cloud computing and artificial intelligence in combination with SD-WAN to reduce false positives, which is an intrusion protection system.

That may be the case. It’s worth noting that software-defined products are not new. Industry experts say that in the 1980s, before microprocessors became powerful enough for widespread use in microcomputers, bit-slicing processors with finite state machines were used on CPU boards. If the team wanted to emulate a DG Nova, they installed an application; if they wanted to emulate a PDP11, they loaded different software.

What shifted the market from dedicated ASICs to microprocessors was the combination of Linux and a new generation of microprocessors that could handle the memory and I/O throughput required for products such as SD-LAN switches, SD Storage, and now SD-WAN. In fact, all of Bridgeworks’ bridging and data acceleration products have been software defined since 2004.

SD-WAN: Huge potential

The potential for SD-WAN products is huge. However, adopters need to get the basics right, and the biggest issue is performance. More and more data is transmitted over longer distances, requiring more and more bandwidth, so if SD-WAN is going to make it, the industry must move up the performance ladder. So, will adding flashy features have an impact on performance? Not likely, at least not in the long run.

AI and AIOp are a focus for many vendors. However, as seen in other AI implementations, a lot of research and thought must go into it. This involves having to think differently.

Sturt commented: “According to Gartner, automation technologies can run more than 70% of network tasks such as adds, moves and changes. This means faster delivery, fewer disruptions and reduced WAN issues. Artificial intelligence for IT operations (AIOps) and DevOps will be able to make changes around the clock.”

Vendors and analysts often talk about how AIOps will revolutionize SD-WAN usage and automate so many functions based on data and patterns. They also say it enables enterprises to create policies for different types of data. However, if they are going to use AI, they should let the technology create the data policy.

Accelerate critical data

AI should also be used to identify data usage patterns and accelerate critical data. There are WAN optimization and data acceleration tools that can have a real impact on data transmission over the WAN (it should be noted that WAN optimization often fails to live up to its promise).

Furthermore, more than 70 years after the first commercial computer, AI should enable self-configuring and self-managing products and must be used to augment and remove human burdens to improve data, task, and network quality, efficiency, and performance.

So, is automation the next stage in leveraging the “extensive insights and data already generated by SD-WAN systems”? Currently, there are two ways to increase data throughput over the WAN:

(1) WAN optimization, which is true data optimization because it uses deduplication to reduce the amount of data transmitted. This is a great technology for file data with repeated characters (such as Microsoft Word files).

(2) WAN Acceleration parallelizes data into many flows controlled by AI. In this case, AI can be self-managed and self-configured, so no supervision is required and enterprises can install it. The advantage of WAN acceleration compared to WAN optimization is that it does not care about the format of the data. It treats all data equally, so it can accelerate already encrypted data. WAN acceleration can be built into higher SD-WAN products as an overlay to support WAN bandwidths of 5Gb/s and above; or it can be located between SD-WAN and WAN.

Improve WAN performance

Enterprises looking to significantly improve their WAN and SDWAN performance will benefit from deploying selective IP port WAN acceleration. It uses data parallelism to increase throughput over the WAN. To maximize the benefits of WAN acceleration, it is important to maintain the maximum data flow into the WAN acceleration device. This is in contrast to WAN optimization, which attempts to optimize data on all IP ports.

For SD-WAN, enterprises should allow the artificial intelligence in the SD-WAN to select the traffic to be accelerated. SD-WAN and WAN acceleration are designed to correspond to each other. By putting them on the same WAN, this can maximize the throughput of the WAN, and the advantage of WAN acceleration is that it is transparent to the SD-WAN.

In situations where you need to send bulk encrypted or compressed data (such as video), it is important to route that data to a WAN acceleration device.
To ensure that they choose an SD-WAN package or device that matches or exceeds the bandwidth capabilities of WAN acceleration, it is critical to ensure that they choose an SD-WAN package or device that matches or exceeds the bandwidth capabilities of WAN acceleration. In situations where bandwidth management is required during the day or night, enterprises can utilize the throttling capabilities in WAN acceleration to increase throughput without disrupting other traffic.

Preventing Data Loss

Security industry experts often mention the benefits of a 3-2-1 backup strategy to prevent data loss: three copies of the same data, two copies on different storage media, and an offsite copy. But what if everything goes wrong and an offsite copy is needed? What if the cloud provider's cloud platform server is also down? With WAN acceleration, further remote replication can be easily carried out thousands of kilometers away without performance degradation caused by latency and packet loss.

Experts say that if you can easily avoid it, don't choose one over the other, and WAN acceleration provides enterprises with 3-2-2 with little added complexity and cost. While WAN acceleration gives enterprises the freedom to eliminate latency and packet loss, it is still critical for air-gapped offsite backups, as cyberattacks now focus first on backup software and files to prevent recovery from backups. Without this security air gap, recovery may never be possible. Therefore, enterprises can undoubtedly benefit from WAN acceleration, but when it comes to network security, other strategies should also be implemented.