Hi, I'm totally new to hadoop, and I was reading the docs on HDFS. It mentions something called "streaming data access". Does this simply mean that HDFS has been optimized for throughput rather than latency? I guess the part I'm having trouble understanding is the "streaming" aspect of it. Except when the amount of data you're dealing with is small enough to allow you to see it all at once, data is always coming to you in the form of a "stream".
daniel
