--- title: "SD-WAN optimization" canonical: "https://docs.aryaka.com/space/KNOW/1283981597/SD-WAN%20optimization" format: markdown --- For mission- and business-critical applications like collaboration tools and big data file systems, enterprises need to use solutions that maximize the efficiency of the data flow across the enterprise WAN. If businesses depend on MPLS for their high-performance WAN needs, bandwidth can still be prohibitively expensive, especially in remote geographies. So enterprises need to look at other ways to conserve and optimize bandwidth. Different vendors have different approaches to achieving that goal, but data reduction technologies and techniques remain at the core of any solution. The principal goal of any WAN optimization solution is to maximize the efficiency of the data flow across the enterprise network. Among the many techniques leveraged in WAN optimization solutions, data deduplication and compression  account for a significant reduction in data flowing across the WAN. This results in huge bandwidth savings, and increases throughput. Traditional WAN optimization vendors use proprietary hardware to attempt to solve the problem. This device-centric approach still relies on the underlying network while requiring massive IT resources for deployment, maintenance, and management. In contrast, Aryaka’s optimization software stack is hosted on hardware managed and maintained by Aryaka Networks, so customers can avoid the hassle. It is built with on-demand application optimization and acceleration that is delivered as a service, and an integral part of our cloud-first WAN approach. Additionally, Aryaka’s multi-tenant proprietary and patented optimization software stack—hosted on every Aryaka POP—combines the power of multi-segment TCP optimization , byte-level compression , and advanced redundancy removal (ARR). These technologies guarantee throughput, induce bandwidth savings, and accelerate all types of applications: on-premises, cloud, and SaaS. Data reduction over Aryaka The following graphic shows the data reduction percentages for many popular protocols using Aryaka’s cloud-first WAN-as-a-service. These protocols include HTTPS/HTTP, IMAPS, MSSQL, CIFS/SMB and applications like Splunk, Citrix, and SAP. All of the following protocols and applications can see substantial bandwidth usage savings using Aryaka. Multi-segment transport and application acceleration over Aryaka  Aryaka's Global SD-WAN architecture is divided into three segments: first, middle (Private L2 and L3 Core), and last mile. Aryaka's optimization techniques, security, routing, and the underlying transport vary for each segment. Each segment of Aryaka's multi-segment architecture is described in a section that follow. Last mile Enterprise customers often adapt and procure direct internet connectivity to connect their datacenters and remote and branch offices to Aryaka’s globally distributed POPs. Procuring, provisioning, managing, and negotiating are some of the challenges faced by an organization's IT teams. To remove some of the challenges, Aryaka offers an optional service to customers known as Aryaka Last Mile Services. This service simplifies contract management and payment processing, and entirely relieves enterprises of basic network responsibilities, including link procurement, deployment and monitoring, and responding to outages.  The following key capabilities are included with Aryaka Last Mile Services: Link procurement—Provides activation planning and procurement of tier 1 broadband services regionally and worldwide. Service contract management—Provides a single Aryaka contact point to manage individual contracts across all vendors. Customer support—Aryaka support agents work as the first point of contact with ISPs when services are disrupted. Their goal is to identify the root cause, improve link quality, and get you up and running fast. Provisioning and testing—Coordinates the provisioning and activation of the internet connection with all of your global service providers. 24/7/365 link monitoring—Provides around-the-clock monitoring of links, packet loss, and latency to proactively address network concerns. Service impact analysis—Produce in-depth data analysis from link monitoring, service degradation, and service disruptions to prevent future challenges and outages. Customer notifications—Send regular updates on service issues, detailed incident reports, monthly summary reports, and upcoming maintenance activities. Maintenance scheduling—Oversee, coordinate, and review all planned and unplanned ISP maintenance activities. Middle mile Aryaka's services offer enterprises the ability to connect their datacenters, remote and branch offices to Aryaka's global private L2/L3 full mesh network. This network offers various optimization technologies for your users accessing enterprise applications and and all other traffic requirements. The following key capabilities are included in the middle mile: Globally distributed POPs that are interconnected by dedicated Layer 2 links. All POPs are fully redundant and are located within 30 milliseconds of any knowledge worker across the globe. Full mesh architecture ensures end-to-end application performance while delivering the industry's best SLAs. First mile The first mile connection provides end-to-end connectivity utilizing various connection methods directly from one of the Aryaka's globally-distributed POPs. This connectivity is achieved using one of the following three scenarios based upon the link connectivity: From an Aryaka POP to IaaS over private L2 connections. From an Aryaka POP to IaaS over internet L3 connections. From an Aryaka POP to SaaS over internet L3 connections. The following key capabilities are included in the first mile: Cloud acceleration: Allows customers to connect to their SaaS applications over an SLA-driven connection from the nearest Aryaka POP. Ensures low latency and reduces packet loss issues when internet is used as an enterprise's middle-mile. Cloud transport gateway: IaaS WAN gateway integration (for example with Microsoft Azure) ensures that the customer's VNET is available as part of the enterprise WAN. Customer resources deployed in the cloud becomes available to other enterprise sites. Cloud Services Extensions (CSX): Scales the geographic reach and allows for rapid access to IaaS cloud services without the need of a dedicated services POP in a defined region. Service is enhanced with a dedicated interconnections in various regions across the continents. Cloud direct connect: Out-of-the-box connectivity to IaaS providers including AWS, Microsoft Azure, Google, and Alibaba clouds. Directly connect from the closest Aryaka POP to nearest region presence for the IaaS provider. Virtual Appliance (V-ANAP) extends an enterprise's AWS IaaS benefits including Aryaka's optimization technologies end-to-end in the cloud. The three segments are shown in the following graphic: Aryaka optimization technologies This section describes the optimization techniques and technologies incorporated in the Aryaka's SD-WAN solution. TCP and UDP optimization Aryaka uses patented multi-segment TCP and UDP optimization techniques to improve WAN performance. Enterprise cloud applications including Microsoft Office 365, SAP, and Salesforce and unified communications as a service (UCaaS) products deliver a LAN-like experience to users. Aryaka enhances the TCP performance by implementing high and low TCP priorities. The high and low priority queues are only used for accelerated TCP traffic. Aryaka observes all TCP sessions coming from all users that belong to your organization and prioritizes the TCP sessions into high priority sessions and low priority sessions. High priority sessions get processed first based on the priority weight assigned to it.  Class-based WAN rate control (WRC) Aryaka's class-based WRC guarantees fair connection bandwidth control by monitoring end-to-end aggregate utilization for TCP traffic and scaling back bandwidth usage without dropping packets. This helps when multiple TCP connections are competing for the same bandwidth. The following key benefits result from WRC: WRC improves performance when a large number of TCP connections actively transfer data across a link. WRC allows 2000 connections—5% of which are concurrently transferring—to share a 2 mbps link. Tcp connections each increase their individual rate when they experience no loss, and reduce their individual rate when they do experience loss. When a large number of connections share a path, they compete with each other for bandwidth. This results in loss and inefficiency. WRC enforces fair sharing of available bandwidth across all actively transferring connections. WRC improves performance under lossy conditions with the performance improvement higher when latency is higher. WRC improves link utilization to more than 80% of available bandwidth at up to 5% loss, even with up to 100ms edge latency. In lossy conditions, TCP normally decreases and overall throughput is reduced. This gets worse as latency increases. WRC changes TCP behavior such that the aggregate rate grows even at a loss of 5%.  WRC grows rate aggressively up to the slow growth threshold (SGT) , which should be set to the percentage of link available for TCP). Above this threshold, it reduces growth to a very low rate, to avoid negatively impacting non-TCP traffic.        With WRC, we can observe greater than 80% link utilization even on a link showing a 5% loss. Data compression and deduplication Aryaka delivers bandwidth optimization that reduces the file size of data that is transmitted over the network. This optimizes the available bandwidth across the first mile. Link Assure Aryaka offers the Link Assure feature that provides enterprises the benefits of link aggregation and reduced network complexity at branch offices. It includes load balancing, path selection, and single- and dual-link packet loss recovery with proprietary error-correction algorithms. Aryaka QoS details The Quality of Service (QoS) feature of Aryaka's platform enables enterprises to prioritize different types of traffic between multiple locations. Scheduling and queuing Aryaka uses the Hierarchical Token Bucket (HTB) algorithm as part of its QoS implementation. HTB is used to create a hierarchical queue structure and determine relations between queues (priority, burst possibility, and so on). The HTB algorithm also shares the bandwidth between different classes of traffic using a process that allows certain classes to borrow bandwidth from other classes that are not using it. Marking and classification Aryaka honors the marked packets that are coming to the SD-WAN device. Aryaka can also mark or remark packets on the SD_WAN device. Marking and queuing takes place at the ingress of SD-WAN based on 5-tuple flow classification. The following five QoS queues and classes are supported on the ANAP—they are listed in decreasing order of priority: Real Time Mission Critical Transactional Productivity Best Effort The different traffic types are processed as follows: TCP-accelerated traffic is traffic optimized by the Aryaka TCP optimization engine. Accelerated traffic is organized into two classes: Transactional class (high-priority class) Productivity class (low-priority class) Pass-through traffic is non-TCP traffic (for example, UDP, GRE, and ICMP traffic). Pass-through traffic is organized into two classes: Real Time class (high-priority class) Mission Critical class (low-priority class) All other traffic uses the Best Effort class. The following table lists the recommended classification and traffic profiles: Class Name Priority Description Example Real Time (RT) Highest Requires low latency and has low bandwidth Critical voice UDP-based traffic that is latency sensitive is a candidate for high priority pass-through traffic. Mission Critical (MC) Medium Variable rate, variable size Video, voice signaling, and routing protocols are candidates for low priority pass-through traffic. Transactional (TR) Medium Variable rate, short lived elastic flows Interactive database transactions, TCP applications that burst, Citrix VDI, and other applications that require quicker response times (user interaction) are candidates for high priority accelerated traffic. Productivity (PR) Medium High throughput data, bursty traffic Email applications, bulk data and file transfer over TCP/UDP, and sync apps are candidates for low priority accelerated traffic. Best Effort (BE) Lowest All other traffic types, and recreational traffic All other non-critical and recreational applications (for example, YouTube and Facebook). Aryaka offers reserve (guaranteed bandwidth) and limit (maximum bandwidth) per class. For any bandwidth not reserved, the highest class gets priority, followed by the next highest class, and so on (using the following priority order: RT, MC, TR, PR, then BE). Aryaka supports absolute priority between queues and classes. Benefits of Aryaka QoS Enables effective transportation of voice and data traffic Optimizes and prioritizes latency-sensitive applications Limits bandwidth used by non-critical, recreational traffic Adaptive Quality of Service (A-QOS) Aryaka also supports Adaptive QoS for sites with an SD-WAN edge device (that is, an ANAP) used as an edge router. A-QOS can dynamically allocate more bandwidth to the internet when it is not used by the Aryaka network. For example, if Aryaka is using less than the guaranteed bandwidth and there is no network congestion, then Aryaka dynamically increases the internet bandwidth up to the configured link size. However, if traffic is going to the Aryaka network, and there is congestion, then Aryaka uses its guaranteed bandwidth, and internet traffic uses the remaining configured ISP link size bandwidth. When this feature is enabled, A-QOS is configured to know each ISP’s size. It then determines what is available to the internet by calculating what is currently used by the Aryaka network, and then adding a buffer (known as  guardband ) to it. The remaining bandwidth is available for internet. The guardband is added to the Aryaka network utilization to allow the Aryaka network’s utilization to grow quickly to the link bandwidth without having to compete with the internet. SSL and CIFS For applications that use protocols that are latency sensitive, Aryaka leverages application proxy technology with techniques to reduce or mitigate latency impact on the WAN. This provides better performance for WAN-intolerant applications and enhances the end-user experience. In this topic Related topics QoS Link Assure for VPN Path Aryaka Link Assure for VPN Path Internet View optimization rules