Performance tests that measure only abilities to pass packets or open and close connections are simply not enough to evaluate the performance of an ADC. It is necessary to compare ADCs from the perspective of decisions-per-second rather than mere protocol-per-second measures.
Scalability is an important aspect of availability and performance. Scalability, which includes a lot of seemingly unrelated technologies, focuses primarily on increasing available resources to meet demand. Equally important is the ability to failover from one application instance to another or one data centre to another — or to the cloud. Failover capacity is important because it is critical to business continuity.
Failover and scaling are more interrelated than they first appear. These rely on a third capability — visibility — to provide accurate, actionable data upon which an ADC can base its routing decisions, and share with other infrastructure components responsible for failure and application scaling tasks.
The ADC is most often responsible for scaling of applications — whether those applications are deployed in virtual containers or on physical machines. When choosing an ADC, it is important to consider the level of integration and support for various automation, orchestration, and virtualisation solutions, particularly those in the realm of provisioning.
ADCs that easily support heterogeneous environments, bring organisations the benefits of operational consistency such as lower management costs, repeatable deployments across environments, and consistent enforcement of security policies.
Visibility has always been important to application delivery especially as applications become more fluid, multi-tier, multi-server and geographically dispersed.
The ADC combines these multiple sources of information into single application views and intelligently uses the individual status points to control the flow of traffic. It is the means to ensure scalability while at the same time addressing performance and availability issues.
Visibility is also important to automation and integration, enabling the automated scaling out and back down of applications across environments. Visibility is key to detecting attacks and preventing them. In short, visibility is a crucial capability of an ADC that should not be treated as a checkbox item. It should be investigated fully to ensure comprehensive views and functionality.
Because of its location in the data centre architecture, an ADC is uniquely positioned to provide security in a variety of ways to protect not only the application but the computing and network resources upon which applications rely.
Increasingly, this strategic location requires advanced security such as data centre, network, and web application firewall services. As an intermediary between clients and services, an ADC offers a cost-effective and processing-efficient solution for deploying security services.
From the client perspective, an ADC is the endpoint and a more appropriate point for network and data centre firewall services than an upstream or downstream device. Similarly, the ADC must necessarily intercept and examine requests and responses to perform advanced load balancing and application routing. This enables the ADC to examine content in depth and to ensure it is free of virus or malicious code.
Sign up for Computerworld eNewsletters.