Optimization & Acceleration
Ipanema's Optimization & Acceleration features:
Define Application Performance Objectives per user and enforce them globally over the WAN
The Ipanema System's unique ability to guarantee application performance comes from its objective based approach to traffic management. Unlike any other solution on the market, no local device configuration is required to manage the traffic. The objectives are defined in the central management software which then communicates them to the ip|engines. The ip|engines cooperate with each other to form a fully distributed system under the control of the global objectives. In this System there is no single point of failure: if the central management server goes down the System continues to function correctly on the network. This differs from traditional approaches which try to manually configure how traffic should be managed device by device, hoping that good application performance will be the end result.
With Ipanema, enterprises need only to define are which applications matter the most for them and their relative criticalities to the business. A higher criticality is set for more important applications, for which the business needs higher availability. A low criticality does not mean that the application does not matter for the business. It simply means that, when demand for resources is high, the business can tolerate that the application receives less resources.
Applications are defined in the System using identification criteria from network Layer 3 to 7. For example Citrix and HTTP applications can be defined based on published application names and URLs respectively. Once defined, each application has a per-user service level set. This per user service level defines what the network should deliver in terms of network resources for each individual user of a given application. Resources are defined using network attributes such as bandwidth, delay, jitter and loss thresholds. This service level information is contained in a dictionary in the central management software and can be customized to match each enterprise's unique application requirements.
Once the objectives have been defined, the Ipanema System is able first to assess 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 Optimization is activated, the System automatically reapplies the available resources to optimize end-user quality of experience and productivity using the global objectives. In the example below, at site D, critical applications user now receive the required network resources, resulting in these users being 100% satisfied across the entire network.
However, as you can see in the example below, this does not always fix all the performance problems in a network. Through Acceleration, Ipanema is able to mitigate the impact of delay and increase the bandwidth available for applications. This means that the business critical application users are be satisfied everywhere. For low criticality or recreational applications, enterprises can make the decision to voluntarily limit their performance to discourage their use.
Accelerate while Guaranteeing critical application performance
Acceleration brings response time down, facilitating the quality of experience that users expect. A combination of redundancy elimination, TCP acceleration and application acceleration mechanisms are dynamically applied under the control of Ipanema’s sophisticated global, dynamic optimization features.
Being an Autonomic Networking System, the mechanisms are triggered dynamically depending on the nature of the applications. The system automatically adapts the application flows to emerging conditions in real-time based on a dynamic, global analysis of the traffic mix, user behavior and continuous tracking of WAN resource availability.
Ipanema’s Acceleration is:
- Controlled: Acceleration features are tightly coupled with Ipanema’s Optimization.
Ipanema Optimization prevents Acceleration from creating congestion – even in meshed networks
The “virtual bandwidth” created by Acceleration is allocated through Optimization, i.e. according to business priorities.
- Transparent to the IT infrastructure: Ipanema’s Acceleration does not require any change to the configuration of the IT infrastructure(and vice-versa)
- Transparent to the Network: Ipanema’s Acceleration only modifies the payload of the packets, it preservers layer 3 and 4 information and does not use tunnels.
- Able to manage all applications: Acceleration is complete and covers the three key application performance bottlenecks (bandwidth, TCP itself and the application protocol) - together with Optimization it addresses all types of applications: real-time, transactional and data-transfers.
- A la carte: Individual acceleration mechanisms can be deployed only where they are really required - A branch device is not always required to get the benefit of Intelligent Acceleration
Globally manage meshed flows from the datacenter
In modern networks, competition for network resources occurs not only among applications but also among sites. In this example, a branch office accesses a critical SAP application over the WAN from the main data center. At the same time, the same user’s email application synchronizes data over the WAN from regional a data center. Then another user in the same branch also generates email traffic from the main data center. The resulting competition between application flows from the data centers creates congestion at the PE router at the branch, resulting in poor performance of SAP. Although such competition can be handled with a traffic management device located in the branch as long as the traffic from data center does not contain non-TCP traffic), controlling it on the destination side is not optimal.
Through the global management of network traffic, made possible by Cooperative Tele-Optimization between the devices located at the data centers, the congestion in the PE router at the branch can be completely avoided, even without installing a device there. The cooperating ip|engines in the datacenters exchange information in real-time about the flows they are controlling, and from that co-ordination they detect that they are both sending traffic to the branch. They dynamically compute the bandwidth that should be given to each user session going to the branch based on their shared knowledge of the traffic mix, its business criticality and the resources available. This prevents any congestion from happening in the destination PE router by controlling the traffic at its source.
Cooperative Tele-Optimization dramatically reduces the costs and the complexity of application traffic management over large distributed networks by removing the need for appliances in all branches.
Dynamically protect interactive applications and enable convergence
The Ipanema System optimizes more than just bandwidth – it also manages other characteristics of network resources: delay, loss and jitter. Smart Packet Forwarding technology allows application flows to be managed differently depending on their requirements and on individual user behavior. Packets belonging to real-time flows, such as Voice over IP or video, are forwarded in a manner that prevents the injection of unwanted delay, loss or jitter.
Packets belonging to data-transfer flows, such as with FTP or Email, are forwarded so that they receive the appropriate bandwidth resources without degrading other more delay sensitive flows. Packets belonging to interactive flows, such as with Citrix or Windows RDP, typically require the lowest transit delay. These flows are analyzed by Ipanema to detect the user behavior in real-time. This unique ability is required to deal with “hybrid” flows such those where users load or save a local file inside a Citrix session, such as document editing. Ipanema is able to prevent the data-transfer phases of the interactive flow from freezing other user’s interactive Citrix sessions.
Combine multiple physical networks into a unified logical network
Smart Path Selection maximizes the value of multiple network-access-path implementations by unifying the different accesses at the global enterprise level and ensuring the optimum usage of each available path between sites. It dynamically selects the best network access path for each session in order to guarantee the associated application SLA and maximize the application performance and security.
Smart Path Selection provides Enterprises with full business communication continuity through redundant networks while guaranteeing application performance and optimizing their network costs.
Tightly integrated with the other Ipanema features of Visibility, Optimization and Acceleration, Smart Path Selection delivers benefits over redundant networks, with no extra operational workload.
Smart Path Selection:
- Dynamically selects the best network access path for each session maximizing application performance, security and network usage.
- Is built on the Ipanema System’s ability to enforce Application level SLAs guaranteeing the performance of all applications on the network (data, voice, video, interactive …).
- Makes fast decisions based on real-time path availability and quality, to prevent link degradations and failures from having a noticeable impact on application performance and user quality of experience.
- Makes all resources always ready for use (vs. standard link failover), to ensure the best possible usage of existing network resources.
