Claims of speed hikes of up to 1,000 per cent are being made by developers Linus Torvalds and Andrew Morton, both of the Open Software Development Lab, for the Linux 2.6 kernel.
You could be forgiven for being sceptical. To be fair, though, it's also been reported that the main claim for the kernel's improved I/O scheduling can increase database workload processing by as much as 15 per cent.
The improvements are the result of minimising disk head movement during concurrent reads, according to reports. Linux founder Torvalds reportedly said: "Basically, the anticipatory scheduler maintains some per-process statistics to try to guess whether there will be another dependent read 'soon'" and, if the algorithm guesses that there will be another read access, it will wait before returning to process the next request.
"This significantly decreases the back and forth seeking under some loads," the report continues. "The wait-before-seek enables one process to perform several similar reads in rapid succession without having to send the drive head back and forth repeatedly. The anticipatory scheduler is best suited to standard desktop and file server loads. In fact, it was while using the anticipatory scheduling that Morton experienced up to 1,000 per cent speed increases on some workloads."
The second, Deadline Scheduler, adds two queues, FIFO read and FIFO write and adds the claimed "1,000 per cent" performance only on desktop-type loads. Database loads experience improvements of up to 15 per cent, according to developer Andrew Morton. "The deadline scheduler is faster for the short-and-seeky style database workloads, sometimes by up to 15 percent," he said.
The end result is to make desktop applications more responsive to end users because they don't have to wait as long in the queue for attention from the CPU.