Part 2: Finding latency in financial transactions, where every microsecond counts
Component selection and characterization testing
This blog is a continuation of Part 1: The latency problem
Verifying network equipment specifications
Component selection is a critical part of data center design. The specifications of networking, computing, and storage components need to be verified – those that come from the equipment manufacturers are the best possible values and may not be realistic. To establish the proper performance parameters that relate to a site’s activity, you’ll need to test all components with the type of network traffic and applications that will run on the component.
For example, an application delivery controller (ADC) must be tested with a mix of real-world traffic that mimics the traffic the ADC will experience when deployed. This provides a true measure of capacity that can then be matched with other components. In the case of an ADC, the number of servers per ADC can be accurately determined so that the ADC is not limiting the overall system performance.
Testing interoperability and conformance
Component selection testing is also essential in ensuring interoperability between components. Although networking components from different vendors have become relatively interchangeable, there’s always a possibility that networking protocols are implemented differently. For example, the TRILL protocol is a relatively new protocol for optimizing networking paths with a data center. As the protocol matures, there are opportunities for implementation differences by different vendors. Conformance testing, as it’s called, confirms that network devices adhere to published standards.
Look for the upcoming part 3: “Pre-deployment testing”.
For more information
Part 1: The latency problem of Finding latency in financial transactions, where every microsecond counts
See our testing solutions for finance web page
See our Ultra Low Latency (ULL) Testing Blackbook