Date: Sun, 8 Oct 1995 10:26:19 +1000 From: Bruce Evans <bde@zeta.org.au> To: bde@zeta.org.au, se@zpr.uni-koeln.de Cc: hackers@freebsd.org Subject: Re: VLB Disk Controllers Message-ID: <199510080026.KAA08418@godzilla.zeta.org.au>
next in thread | raw e-mail | index | archive | help
>Well, the machine is more severely affected by a PIO >device, than the CPU load indicates. The CPU must be >able to respond to data becoming available within a >very short time, or data may be lost. (Hmm, maybe the >larger buffers of current IDE drives make this a non- >issue ???) PIO is more bursty. Consider transferring 64K at 2MB/sec. That's 32 ms (more than 3 clock ticks) during which softclock() won't run, so timeout functions won't get called. Most interrupts had better have higher priority than PIO devices, and 2 hoggish PIO devices are likely to block each other for too long. This could be handled by time slicing the interrupt handlers. Data loss is only a serious problem for volatile input. Input from drives is nonvolatile and anyway a mere 64K cache would probably be large enough for input at a few MB/sec (it's serial chips with 1 byte fifos at that lose data :-). >Anyway, just a data point: > http://www-itg.lbl.gov/ISS/performance.ps >contains SCSI controller performance data, and they >get 24.5MB/s simultaneously reading from 6 drives >connected to 3 NCR 53c810 controllers in pairs. >The system is reported to be 70% idle under that load >(it's a Pentium 100 with Triton chip set). EIDE certainly can't compete with that. At best it will achieve 15MB/sec throughput on a Pentium 1e6 :-). Bruce
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199510080026.KAA08418>