From cbmorrey at yahoo.com Thu Jan 5 21:46:44 2006 From: cbmorrey at yahoo.com (cb morrey) Date: Thu, 5 Jan 2006 18:46:44 -0800 (PST) Subject: [Disksim-users] Newer Disk Models/diskspecs for Disksim 3.0 Message-ID: <20060106024644.32478.qmail@web34601.mail.mud.yahoo.com> Hello disksim users/developers! I have disksim 3.0 up and running and am adding a new type of cache to the front end of it to test my PhD thesis. I was wondering if there are any newer disk models and diskspecs available than the Quantum Atlas 10K from 1999? I realize that there is a description for automatically profiling disks from userland in Linux called DIXTRAC on the PDL website. I'd be happy to profile my own disk setups if I had access to that software, but I'd be perfectly satisfied with some specs for a recent drive. Any chance of that? Response by private email is fine. Charles Morrey PhD Student Computer Science Department University of Colorado, Boulder (firstname dot lastname at colorado dot edu) __________________________________________ Yahoo! DSL ? Something to write home about. Just $16.99/mo. or less. dsl.yahoo.com From bucy at ece.cmu.edu Thu Jan 5 22:06:40 2006 From: bucy at ece.cmu.edu (John S. Bucy) Date: Thu, 5 Jan 2006 22:06:40 -0500 Subject: [Disksim-users] Newer Disk Models/diskspecs for Disksim 3.0 In-Reply-To: <20060106024644.32478.qmail@web34601.mail.mud.yahoo.com> References: <20060106024644.32478.qmail@web34601.mail.mud.yahoo.com> Message-ID: <20060106030640.GA28446@bismuth.club.cc.cmu.edu> On Thu, Jan 05, 2006 at 06:46:44PM -0800, cb morrey wrote: > Hello disksim users/developers! > > I have disksim 3.0 up and running and am adding a new > type of cache to the front end of it to test my PhD > thesis. I was wondering if there are any newer disk > models and diskspecs available than the Quantum Atlas > 10K from 1999? We're working on a new release which has a new layout model that supports newer disks. We have quite good models of e.g. the Seagate Cheetah15k4. > I realize that there is a description for > automatically profiling disks from userland in Linux > called DIXTRAC on the PDL website. I'd be happy to > profile my own disk setups if I had access to that > software, but I'd be perfectly satisfied with some > specs for a recent drive. Any chance of that? > Response by private email is fine. Dixtrac will be included in the release. john From ozlemakp at gmail.com Mon Jan 23 16:47:03 2006 From: ozlemakp at gmail.com (Ozlem Akpinar) Date: Mon, 23 Jan 2006 16:47:03 -0500 Subject: [Disksim-users] Disk Scheduling Algorithms Message-ID: Hi, Is there a brief description of the 27 scheduling algorithms used by DiskSim? I don't understand what the extensions CYL, and OPT refer to. What is CYCLE_LBN, WPTF_OPT (Weighted with respect to what?) SDF TSPS SPTF_ROT_OPT SPTF_ROT_WEIGHT SPTF_SEEK_WEIGHT Lastly, are C-SCAN, LOOK and C-LOOK simulated? Thank you. Ozlem -------------- next part -------------- An HTML attachment was scrubbed... URL: From bucy at ece.cmu.edu Tue Jan 24 13:45:23 2006 From: bucy at ece.cmu.edu (John Bucy) Date: Tue, 24 Jan 2006 10:45:23 -0800 Subject: [Disksim-users] Disk Scheduling Algorithms In-Reply-To: References: Message-ID: <4fc1e0430601241045n563c3b1p58ca11bed616458b@mail.gmail.com> The manual has references to the various literature where these algorithms are described. And you can always look at the source code. Beyond that, I don't really know. I imagine many of those were "one-offs" that someone needed at some time or another for one experiment that accumulated in the codebase. I'd stick to the algorithms used in validated diskspecs or make sure you really understand what the algorithm is doing before you mess with some of the more esoteric ones. john On 1/23/06, Ozlem Akpinar wrote: > Hi, > Is there a brief description of the 27 scheduling algorithms used by > DiskSim? I don't understand what the extensions CYL, and OPT refer to. > What is > CYCLE_LBN, > WPTF_OPT (Weighted with respect to what?) > SDF > TSPS > SPTF_ROT_OPT > > SPTF_ROT_WEIGHT > > SPTF_SEEK_WEIGHT > > Lastly, are C-SCAN, LOOK and C-LOOK simulated? > > Thank you. > Ozlem > _______________________________________________ > Disksim-users mailing list > Disksim-users at ece.cmu.edu > https://sos.ece.cmu.edu/mailman/listinfo/disksim-users > > > From shahrukh at cs.virginia.edu Fri Jan 27 14:13:35 2006 From: shahrukh at cs.virginia.edu (Shahrukh Rohinton Tarapore) Date: Fri, 27 Jan 2006 14:13:35 -0500 Subject: [Disksim-users] ioreq_event's opid field Message-ID: <43DA70DF.8010401@cs.virginia.edu> Can someone explain the use of opid in the ioreq_event structure? Almost all of the source code sets it to zero and the documentation says to set it to zero when dealing with iotraces. My intention would be to use it as a unique identifier for requests. I want to make sure assigning this variable arbitrary values won't kill the simulator. -- shahrukh From xyao at cse.unl.edu Fri Jan 27 14:26:46 2006 From: xyao at cse.unl.edu (xiaoyu yao) Date: Fri, 27 Jan 2006 13:26:46 -0600 Subject: [Disksim-users] ioreq_event's opid field In-Reply-To: <43DA70DF.8010401@cs.virginia.edu> Message-ID: <000801c62377$a011cf80$2bb05d81@Safari> Hi, Shahrukh It is used for identify a logical request during the request mapping at RAID (or logorg) level. Check disksim_logorg.c and you will find that for each logorg, the opid is maintained like a sequence no. int logorg_maprequest (logorg **logorgs, int numlogorgs, ioreq_event *curr) { ...... curr->opid = logorgs[logorgno]->opid; ...... logorgs[logorgno]->opid++; } Xiaoyu -----Original Message----- From: disksim-users-bounces at ece.cmu.edu [mailto:disksim-users-bounces at ece.cmu.edu] On Behalf Of Shahrukh Rohinton Tarapore Sent: Friday, January 27, 2006 1:14 PM To: disksim-users at ece.cmu.edu Subject: [Disksim-users] ioreq_event's opid field Can someone explain the use of opid in the ioreq_event structure? Almost all of the source code sets it to zero and the documentation says to set it to zero when dealing with iotraces. My intention would be to use it as a unique identifier for requests. I want to make sure assigning this variable arbitrary values won't kill the simulator. -- shahrukh _______________________________________________ Disksim-users mailing list Disksim-users at ece.cmu.edu https://sos.ece.cmu.edu/mailman/listinfo/disksim-users