Storage validation test procedure

All tests are conducted using a Matrox Axio system but the results are also applicable to Matrox RT.X2 systems. A mix of compressed and uncompressed HD and SD footage is used so that the throughput of the storage subsystem can be fully assessed. Using only compressed footage would often require more layers than the editing system could decompress simultaneously.

Four different tests are performed to simulate real-world uses of the storage:

• Maximum throughput represents the best-case scenario. In this test, the playback throughput is incrementally increased until the storage can no longer meet the demands of the editing application.
• Burst throughput demonstrates the maximum sustainable throughput that can be delivered suddenly, with no warning. This test simulates editing several layers from the very beginning of a project. Several video layers all aligned at the very start of the timeline, the 0-second mark.
• Offset burst throughput is similar to burst throughput except that the layers are aligned at the 1-second mark, in other words, the burst occurs with a 1-second warning. This will usually allow the editing application to read ahead and predict the �burst� that will occur at the 1-second mark which will often increase performance. You can actually use this technique in your projects.
• Aggressive throughput is the most challenging test. It is similar to burst throughput except the layers are strategically chosen from very different physical locations on the storage unit. The goal is to test how the seeking characteristics of the storage unit affect performance. Often throughput drops dramatically, similar to the way fragmentation affects a disk.