|Overall Statistics||Video Clip Transcoding|
|Data Read||858.09 MB|
|Data Written||238.46 MB|
|Disk Busy Time||3.07 s|
|Average Data Rate||357.28 MB/s|
According to Cisco, global Internet video traffic surpassed global peer-to-peer (P2P) traffic in 2010, and by 2012, Internet video will account for over 50 percent of consumer Internet traffic. Amazingly, that number does not include the amount of video exchanged through P2P file sharing.
This underscores the pervasiveness of streaming video. Here at Tom's Hardware, we often need to throw technology-related content onto YouTube for reviews and news posts. For others, this task may be more personal than professional. Either way, it's clear that content creation is a daily fact of life.
That's certainly the logic behind Intel's decision to dedicate a block of silicon on its Sandy Bridge-based processors to accelerating video encode and decode. Dubbed QuickSync, the company makes it possible to transcode an entire unprotected Blu-ray movie in under 20 minutes. We already took a very thorough look at output quality and found Quick Sync to be surprisingly good. For more information, read Video Transcoding Examined: AMD, Intel, And Nvidia In-Depth.
In our trace, we transcode an unprotected Blu-ray video clip using the default 720p profile for the Apple iPad with CyberLink's MediaEspresso. We didn't record the time it took to load the application, but we did record 858 MB worth of reads (158 MB more than the size of our original video file). The extra data is attributable to the application's libraries and various other file dependencies for the transcoding task.
We see higher queue depths, though fewer than 1.6% of all operations occur above a queue depth of 32. Even then, we never see any operation queued more than 62 deep. As for transfer size and seek distance, our trace reflects what you'd expect of a video oriented workload: 128 KB transfers accessed sequentially.
- 30% of all operations occur at a queue depth of one
- 56% of all operations occur between a queue depth of two and six
- 91% of all data transferred is sequential
- 75% of all operations are 128 KB in transfer size