The Complete Fair Queuing (CFQ) I/O Scheduler works to achieve similar goals, albeit via a different approach.* With CFQ, each process is assigned its own queue, and each queue is assigned a timeslice. The I/O scheduler visits each queue in a round-robin fashion, servicing requests from the queue until the queue’s timeslice is exhausted, or until no more requests remain. In the latter case, the CFQ I/O Scheduler will then sit idle for a brief period—by default, 10 ms—waiting for a new request on the queue. If the anticipation pays off, the I/O scheduler avoids seeking. If not, the waiting was in vain, and the scheduler moves on to the next process’ queue.
Within each process’ queue, synchronized requests (such as reads) are given priority over nonsynchronized requests. In this manner, CFQ favors reads and prevents the writes-starving-reads problem. Because of the per-process queue setup, the CFQ I/O Scheduler is fair to all processes, while still providing good global performance.
The CFQ I/O Scheduler is well suited to most workloads, and makes an excellent first choice.
The Noop I/O Scheduler
The Noop I/O Scheduler is the most basic of the available schedulers. It performs no sorting whatsoever, only basic merging. It is used for specialized devices that do not require (or that perform) their own request sorting.
