 |
 |
Peter Elmer said... >> Those numbers do have a tendency to confuse people, however. Andy or >> Bill >>Weeks can explain better what they were trying to do, but I understood it >>to >>be (roughly) an attempt to isolate and measure all pieces of the process >>of >>redirecting and opening files in order to insure that each piece within >>the xrootd daemon itself was scalable and performant. This means (again, I >>think) that they were probably measuring this with memory mapped files, >>which >>removes the latencies for finding/opening something on the disk system >>itself >>from their measurement. >> Andy and Bill, is this correct? No. The timing measures how long it takes to open a *disk* file. One has to do that whether or not one memory maps the file. While the times look good they are two orders of magnitude worse than locally opening the file. This doesn't really matter as long as you read most of the file. But still, I would have expected a single order of magnitude. Something to look in to as to where all the time is going. Andy