QoS & Control

Global QoS and control guarantees user experience

Ipanema's QoS and Control combines with Application Visibility, WAN Optimization and Dynamic WAN Selection. All are tightly coupled in the patented Ipanema autonomic networking architecture. This tight coupling in a single coherent system delivers truly optimal results - meaning the best application performance at the lowest cost. Ipanema's approach to QoS and control guarantees a superior user experience and business application acceleration. With the Ipanema System, you can:


Drive performance over the global WAN from application performance objectives

APOs set per-user minimum quality expectations for each important application in the enterprise. The APOs are defined from the central management software (IMSS or SALSA) which then communicates them to the distributed devices (ip|engines).

The ip|engines cooperate with each other in a fully distributed manner in order to enforce the global performance objectives. Enterprises need only to define which applications matter most by their relative criticalities to the business. A higher criticality is set for more important applications where the business needs guaranteed performance. A low criticality does not mean that the application does not matter, only that when demand for resources is high the business can tolerate having users of the lower criticality application receive fewer resources.

Applications are defined using layer 3 to 7 criteria. For example, Citrix and HTTP applications can be defined based on published application names and URLs, respectively. Then a service level is configured for each application. It defines what performance the “optimized network” should deliver for each active user.

Manage the network cost / application performance tradeoff

Once the APOs are defined, the Ipanema System assesses whether or not they are currently being delivered over the network. In the example below, site D has serious application performance problems, including poor performance of high criticality applications such as SAP. When Ipanema QoS and control is activated, the system automatically allocates the available resources to match or even exceed the global objectives. Critical applications at site D then receive the required network resources, resulting in these users being 100% satisfied across the entire network.

APOs can also be used for security reasons or traffic denial. For example, an objective set to 0 Kbits/s will shut that application down, network-wide.

Visibilility + Guarantee + Acceleration  


Protect real-time and interactive flows and converge all application on a single network

Ipanema manages all characteristics of networking, such as delay, loss and jitter. Application flows are handled differently depending on their technical requirements and individual user behavior. For example:

  • Packets belonging to real-time flows such as VoIP or tele-presence are forwarded in a manner that prevents unwanted delay, loss or jitter.
  • Packets belonging to interactive flows such as Citrix, VMware View or Windows RDP typically require low transit delay. They are analyzed to detect user behavior in real-time and manage changing flows, such as where users load or save a local file inside a Citrix session. Ipanema is then able to prevent the data-transfer phases of the interactive flow from freezing other users' interactive Citrix sessions.
  • Packets belonging to data-transfer flows, such as FTP or email, are forwarded so that they receive the maximum possible bandwidth resources without degrading delay-sensitive flows.


Manage globally meshed flows from central sites

Competition for network resources occurs not only among applications but also among sites. In the example below, a user at a branch office accesses a critical SAP application from the main data center over the WAN. At the same time, the user’s email application synchronizes data over the WAN from a regional datacenter. The resulting competition between application flows from the two datacenters creates congestion at the PE router serving the branch, resulting in poor SAP performance.

Ipanema solves this situation (and many that are much more complex) by having a physical ip|engine device at each datacenter and using telemanagement of the branch via a virtual ip|engine.

The cooperating ip|engines exchange real-time information and detect that they are both sending traffic to the branch, creating a painful congestion for SAP. They dynamically compute and apply the resource that should be given to each active user based on their shared knowledge of the traffic mix, business-criticality of each application and the resources available. The congestion is then removed from the destination PE router and user Quality of Experience is excellent.

Manage globally meshed flows from central sites


TOP