Mailing List ArchiveSupport open source code!
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]Re: IDE vs SCSI for RAID
- To: tlug@example.com
- Subject: Re: IDE vs SCSI for RAID
- From: Tobias Diedrich <ranma@example.com>
- Date: Sat, 18 Aug 2001 14:17:15 +0200
- Content-Disposition: inline
- Content-Type: text/plain; charset=us-ascii
- Delivered-To: tlug@example.com
- Fortune: Someone is speaking well of you.
- In-Reply-To: <15229.63895.947565.503292@example.com>
- List-Help: <mailto:tlug-request@example.comsubject=help>
- List-Post: <mailto:tlug@example.com>
- List-Subscribe: <mailto:tlug-request@example.comsubject=subscribe>
- List-Unsubscribe: <mailto:tlug-request@example.comsubject=unsubscribe>
- Mail-Followup-To: Tobias Diedrich <ranma@example.com>, tlug@example.com
- Old-Return-Path: <ranma@example.com>
- References: <F124FRaqbC8wSA1XvpF000090ae@example.com> <15227.39771.9000.307909@example.com> <3B7BA029.80DFE032@example.com> <15228.32466.470584.418429@example.com> <20010817151011.F11053@example.com> <15227.39771.9000.307909@example.com> <3B7BA029.80DFE032@example.com> <15228.32466.470584.418429@example.com> <20010817151011.F11053@example.com> <15229.63895.947565.503292@example.com>
- Reply-To: tlug@example.com
- Resent-From: tlug@example.com
- Resent-Message-ID: <Sjsx5D.A.ZmC.8ylf7@example.com>
- Resent-Sender: tlug-request@example.com
- Sender: Tobias Diedrich <ranma@example.com>
- User-Agent: Mutt/1.3.20i
Stephen J. Turnbull wrote: > Tobias> You did enable Busmaster DMA transfers using hdparm -d1, > > Not on Linux in mid-1996 I didn't. You did notice the 1.2.13, didn't you? Ah, I somehow did forget about that ^^; > Tobias> In contrast to that I get only 20MB/sec > > But who cares about those numbers? (That's a real question: "who" = > "people involved in supporting Application X".) _I_ don't care about > burst transfer rate, which is what "high-performance" IDE is optimized > for (because it's an easily measurable damned lie), and what hdparm > measured the last time I looked (years ago). It may be relevant to > Jc's server application, if what is important is getting longish video > clips to a few dozen humans' workstations in real time. Of course you are not going to notice an increase from 20MB/sec to 25M/sec in bulk transfer speed much except in some special loads, but if the tranfer rate is greater and the seek time is smaller too, an overall better interactive performance should be there. Not always much, but still there. A bit different is the enable_dma parameter of hdparm because it takes some load of the processor and increases interactive performance noticably. > If it's supporting a dozen programmers all working on separate modules > of a large C app (or several of them), it'll be spraying hundreds or > thousands bursts of 1-50 KILObytes, not multiple MB, in short periods > as they rebuild. And if you're into swap (and who isn't?) you're > talking pages (4kB, IIRC)! SCSI is a robust solution that doesn't > require lots of tuning by the admin in that context. SCSI is technically superior in that you can attach 15(7) devices to the host controller and those devices can even "talk directly with each other" (which is rarely used AFAIK). IDE was built to be cheap and is nowadays just as fast as SCSI (If you connect equally fast disks) but because of compatibility issues with older disk driver dma was disabled by default. > Not to run down Jc, but if he needs to ask "what are the advantages of > SCSI?", is he likely to be competent to fine-tune the disk to his > application _now_? I know I'm not. He (and I) could _become_ > competent, I'm sure---but which is cheaper, a SCSI system or our time? Well the only tuning really needed is setting the enable_dma option. (Because that one makes the really big perceivable difference) > The historical answer is "hardware is cheap, wetware is dear." Hmm, if I have to decide wether to spend x $CURRENCY on a drive and spend 5minutes to modify my boot scripts to execute one additional command or 2*x $CURRENCY, then I'm going for the former. But of course you have to know about that first... -- Tobias PGP-Key: 0x9AC7E0BC
- Follow-Ups:
- Re: IDE vs SCSI for RAID
- From: "Stephen J. Turnbull" <turnbull@example.com>
- References:
- IDE vs SCSI for RAID
- From: "Jean-Christian Imbeault" <jean_christian@example.com>
- IDE vs SCSI for RAID
- From: "Stephen J. Turnbull" <turnbull@example.com>
- Re: IDE vs SCSI for RAID
- From: "A.Sajjad Zaidi" <sajjad@example.com>
- Re: IDE vs SCSI for RAID
- From: "Stephen J. Turnbull" <turnbull@example.com>
- Re: IDE vs SCSI for RAID
- From: Tobias Diedrich <ranma@example.com>
- Re: IDE vs SCSI for RAID
- From: "Stephen J. Turnbull" <turnbull@example.com>
Home | Main Index | Thread Index
- Prev by Date: TLUG Tsukuba Branch nomikai
- Next by Date: Re: ipchains question
- Prev by thread: Re: IDE vs SCSI for RAID
- Next by thread: Re: IDE vs SCSI for RAID
- Index(es):
Home Page Mailing List Linux and Japan TLUG Members Links