From roaslin at gmail.com Mon Mar 2 06:35:16 2009 From: roaslin at gmail.com (Raul) Date: Mon, 2 Mar 2009 12:35:16 +0100 Subject: [Disksim-users] Cachemem Message-ID: <5F2196B1-FA2D-467D-BEB6-73EB2A6227A2@gmail.com> Hello, Well, I solved the problems with the ascii traces :) . Now, I want to include Cache in this tests. Do I need to change the bus type?? I use a trace that makes 100 requests to the same block, but I didn?t see any changes in the *.outv. these are the results without cache: Disk number of buffer accesses : 99 Disk buffer hit ratio : 93 Disk buffer miss ratio : 6 I think It is wrong because is the same all the time. Can you help me? thank you From g.kaisling at gmail.com Mon Mar 2 08:10:55 2009 From: g.kaisling at gmail.com (GEOFFREY KAISLING) Date: Mon, 2 Mar 2009 22:10:55 +0900 Subject: [Disksim-users] about the Traces Format Message-ID: Dear all, I intended to use Cello-99 traces as the DiskSim traces. But the traces are the srt, fs and ps format. How to convert them to the suitable trace format which the DiskSim can work? I have tried to adding suppoort for new trace formats in "disksim_iotrace.c", but I don't know how to do it? Would you please tell me how to ? Thank you very much! G.Kaisling -------------- next part -------------- An HTML attachment was scrubbed... URL: From guyh at iis.u-tokyo.ac.jp Wed Mar 4 07:52:15 2009 From: guyh at iis.u-tokyo.ac.jp (Yanhui-iis) Date: Wed, 4 Mar 2009 21:52:15 +0900 Subject: [Disksim-users] about Fig 2 in Page 34 of DiskSim V4.0 Reference Manual Message-ID: <000001c99cc8$0d48b1f0$27da15d0$@u-tokyo.ac.jp> Dear all, In Page 34 of DiskSim V4.0 Reference Manual, an example topology specification is provided in Fig.1, and Fig.2 illustrate the diagram of the storage system corresponding to the specification of Fig.1. I wonder why ctlr0 is at the top of bus0? (I think bus0 should be at the top of ctrl0) Thanks in advance. Best, Yanhui-IIS From sdmcclane at gmail.com Mon Mar 16 12:54:43 2009 From: sdmcclane at gmail.com (Seamus McClane) Date: Mon, 16 Mar 2009 10:54:43 -0600 Subject: [Disksim-users] Compiling Disksim 3.0 on 32-bit systems with gcc 4.1.2 Message-ID: Hi, I am compiling Disksim 3.0 with GCC 4.1.2 on my 32-bit systems (IA32 and AMD32) running CentOS 5.2 Note: I implemented a scheduling algorithm on Disksim 3.0. That's why I need to stay with it, at least for now. Does anyone have a clue? Please help. I got a lot of error messages in the screen output. I grep'ed them and pasted as below. I got exact the same screen output on both IA32 and AMD32 systems. util.c:241: error: label at end of compound statement util.c:829: error: label at end of compound statement make[1]: *** [util.o] Error 1 layout_g1.c:212: error: label at end of compound statement make[1]: *** [layout_g1.o] Error 1 config.h:34:31: error: libparam/libparam.h: No such file or directory config.h:35:32: error: libparam/bitvector.h: No such file or directory modules/disksim_bus_param.h:22: error: array type has incomplete element type modules/disksim_bus_param.h:23: error: ?I? undeclared here (not in a function) modules/disksim_bus_param.h:25: error: ?D? undeclared here (not in a function) modules/disksim_bus_param.h:32: error: variable ?disksim_bus_mod? has initialize r but incomplete type modules/disksim_bus_param.h:32: error: ?lp_modloader_t? undeclared here (not in a function) modules/disksim_bus_param.h:32: error: expected ?}? before ?disksim_bus_loadpara ms? modules/disksim_bus_stats_param.h:18: error: array type has incomplete element t ype modules/disksim_bus_stats_param.h:24: error: variable ?disksim_bus_stats_mod? ha s initializer but incomplete type modules/disksim_bus_stats_param.h:24: error: expected ?}? before ?disksim_bus_st ats_loadparams? modules/disksim_cachedev_param.h:24: error: array type has incomplete element type modules/disksim_cachedev_param.h:31: error: ?S? undeclared here (not in a functi on) modules/disksim_cachedev_param.h:36: error: variable ?disksim_cachedev_mod? has initializer but incomplete type modules/disksim_cachedev_param.h:36: error: expected ?}? before ?disksim_cachede v_loadparams? modules/disksim_cachemem_param.h:34: error: array type has incomplete element ty pe modules/disksim_cachemem_param.h:36: error: ?LIST? undeclared here (not in a fun ction) modules/disksim_cachemem_param.h:56: error: variable ?disksim_cachemem_mod? has initializer but incomplete type modules/disksim_cachemem_param.h:56: error: expected ?}? before ?disksim_cacheme m_loadparams? modules/disksim_ctlr_param.h:24: error: array type has incomplete element type modules/disksim_ctlr_param.h:30: error: ?BLOCK? undeclared here (not in a functi on) modules/disksim_ctlr_param.h:36: error: variable ?disksim_ctlr_mod? has initiali zer but incomplete type modules/disksim_ctlr_param.h:36: error: expected ?}? before ?disksim_ctlr_loadpa rams? modules/disksim_ctlr_stats_param.h:28: error: array type has incomplete element type modules/disksim_ctlr_stats_param.h:44: error: variable ?disksim_ctlr_stats_mod? has initializer but incomplete type modules/disksim_ctlr_stats_param.h:44: error: expected ?}? before ?disksim_ctlr_ stats_loadparams? modules/disksim_device_param.h:13: error: array type has incomplete element type modules/disksim_device_param.h:17: error: variable ?disksim_device_mod? has init ializer but incomplete type modules/disksim_device_param.h:17: error: expected ?}? before ?disksim_device_lo adparams? modules/disksim_device_stats_param.h:27: error: array type has incomplete elemen t type modules/disksim_device_stats_param.h:42: error: variable ?disksim_device_stats_m od? has initializer but incomplete type modules/disksim_device_stats_param.h:42: error: expected ?}? before ?disksim_dev ice_stats_loadparams? modules/disksim_disk_param.h:81: error: array type has incomplete element type modules/disksim_disk_param.h:150: error: variable ?disksim_disk_mod? has initial izer but incomplete type modules/disksim_disk_param.h:150: error: expected ?}? before ?disksim_disk_loadp arams? modules/disksim_global_param.h:24: error: array type has incomplete element type modules/disksim_global_param.h:36: error: variable ?disksim_global_mod? has init ializer but incomplete type modules/disksim_global_param.h:36: error: expected ?}? before ?disksim_global_lo adparams? modules/disksim_iodriver_param.h:20: error: array type has incomplete element ty pe modules/disksim_iodriver_param.h:28: error: variable ?disksim_iodriver_mod? has initializer but incomplete type modules/disksim_iodriver_param.h:28: error: expected ?}? before ?disksim_iodrive r_loadparams? modules/disksim_iodriver_stats_param.h:27: error: array type has incomplete elem ent type modules/disksim_iodriver_stats_param.h:42: error: variable ?disksim_iodriver_sta ts_mod? has initializer but incomplete type modules/disksim_iodriver_stats_param.h:42: error: expected ?}? before ?disksim_i odriver_stats_loadparams? modules/disksim_iomap_param.h:21: error: array type has incomplete element type modules/disksim_iomap_param.h:30: error: variable ?disksim_iomap_mod? has initia lizer but incomplete type modules/disksim_iomap_param.h:30: error: expected ?}? before ?disksim_iomap_load params? modules/disksim_ioqueue_param.h:29: error: array type has incomplete element type modules/disksim_ioqueue_param.h:46: error: variable ?disksim_ioqueue_mod? has in itializer but incomplete type modules/disksim_ioqueue_param.h:46: error: expected ?}? before ?disksim_ioqueue_ loadparams? modules/disksim_iosim_param.h:18: error: array type has incomplete element type modules/disksim_iosim_param.h:24: error: variable ?disksim_iosim_mod? has initia lizer but incomplete type modules/disksim_iosim_param.h:24: error: expected ?}? before ?disksim_iosim_load params? modules/disksim_logorg_param.h:32: error: array type has incomplete element type modules/disksim_logorg_param.h:52: error: variable ?disksim_logorg_mod? has init ializer but incomplete type modules/disksim_logorg_param.h:52: error: expected ?}? before ?disksim_logorg_lo adparams? modules/disksim_pf_param.h:18: error: array type has incomplete element type modules/disksim_pf_param.h:24: error: variable ?disksim_pf_mod? has initializer but incomplete type modules/disksim_pf_param.h:24: error: expected ?}? before ?disksim_pf_loadparams ? modules/disksim_pf_stats_param.h:20: error: array type has incomplete element type modules/disksim_synthgen_param.h:48: error: variable disksim_synthgen_mod has initializer but incomplete type modules/disksim_synthgen_param.h:48: error: expected } before disksim_synthge n_loadparams modules/disksim_synthio_param.h:22: error: array type has incomplete element typ e modules/disksim_synthio_param.h:32: error: variable disksim_synthio_mod has in itializer but incomplete type modules/disksim_synthio_param.h:32: error: expected } before disksim_synthio_ loadparams disksim.c:490: error: conflicting types for disksim_global_loadparams modules/disksim_global_param.h:8: error: previous declaration of disksim_global _loadparams was here modules/disksim_global_param.c:8: error: paramvec undeclared (first use in thi s function) modules/disksim_global_param.c:8: error: (Each undeclared identifier is reported only once modules/disksim_global_param.c:8: error: for each function it appears in.) modules/disksim_global_param.c:11: error: dereferencing pointer to incomplete ty pe modules/disksim_global_param.c:20: error: dereferencing pointer to incomplete ty pe modules/disksimpe modules/disksim_global_param.c:34: error: dereferencing pointer to incomplete ty pe modules/disksim_global_param.c:38: error: dereferencing pointer to incomplete ty pe modules/disksim_global_param.c:45: error: dereferencing pointer to incomplete ty pe modules/disksim_global_param.c:48: error: dereferencing pointer to incomplete ty pe modules/disksim_global_param.c:51: error: dereferencing pointer to incomplete ty pe modules/disksim_global_param.c:54: error: dereferencing pointer to incomplete ty pe modules/disksim_global_param.c:57: error: dereferencing pointer to incomplete ty pe modules/disksim_global_param.c:60: error: dereferencing pointer to incomplete ty pe modules/disksim_global_param.c:77: error: dereferencing pointer to incomplete ty pe modules/disksim_global_param.c:98: error: dereferencing pointer to incomplete ty pe modules/disksim_global_param.c:111: error: dereferencing pointer to incomplete t ype _global_param.c:25: error: dereferencing pointer to incomplete type modules/disksim_global_param.c:34: error: dereferencing pointer to incomplete ty pe modules/disksim_global_param.c:38: error: dereferencing pointer to incomplete ty pe modules/disksim_global_param.c:45: error: dereferencing pointer to incomplete ty pe modules/disksim_global_param.c:48: error: dereferencing pointer to incomplete ty pe modules/disksim_global_param.c:51: error: dereferencing pointer to incomplete ty pe modules/disksim_global_param.c:54: error: dereferencing pointer to incomplete ty pe modules/disksim_global_param.c:57: error: dereferencing pointer to incomplete ty pe modules/disksim_global_param.c:60: error: dereferencing pointer to incomplete ty pe modules/disksim_global_param.c:77: error: dereferencing pointer to incomplete ty pe modules/disksim_global_param.c:98: error: dereferencing pointer to incomplete ty pe modules/disksim_global_param.c:111: error: dereferencing pointer to incomplete t ype modules/disksim_global_param.c:126: error: dereferencing pointer to incomplete t ype modules/disksim_global_param.c:145: error: dereferencing pointer to incomplete t ype modules/disksim_global_param.c:170: error: missing terminating " character modules/disksim_global_param.c:171: error: missing terminating " character modules/disksim_global_param.c:172: error: ?name? undeclared (first use in this function) make[1]: *** [disksim.o] Error 1 make: *** [all] Error 2 From janak_ce2005 at yahoo.com Mon Mar 16 20:56:07 2009 From: janak_ce2005 at yahoo.com (janak patel) Date: Mon, 16 Mar 2009 17:56:07 -0700 (PDT) Subject: [Disksim-users] Having problem using TPC-C trace with Disksim Message-ID: <526105.81024.qm@web110801.mail.gq1.yahoo.com> Hi, I am using TPC-C trace but I am not able to make it work. LBA in TPC-C trace is very large than the range of the unsigned long int. How can I make this trace environment suitable to simulation environment? I tried changing the value of 'locScale' in I/O mappings block in parameter file but it doesn't work for me. For example: Request 1: 3247.201 0 276996096 3 0 It gives following error: unexpected request location: devno 0, blkno 276996096, bcount 3 Assertion failed: simtime = 3247.201000 totalreqs = 1 disksim: disksim_logorg.c:763: logorg_maprequest: Assertion `logorgno != -1' failed. Aborted Request 2: 5288.554 0 3201773568 16 0 It gives the following error: unexpected request location: devno 0, blkno 2147483647, bcount 16 Assertion failed: simtime = 5288.554000 totalreqs = 1 disksim: disksim_logorg.c:763: logorg_maprequest: Assertion `logorgno != -1' failed. Aborted In the second request, LBA value gets rounded in the range of long int. How can I fix this problem? Thank you! Janak Koshia -------------- next part -------------- An HTML attachment was scrubbed... URL: From stroucki at ece.cmu.edu Mon Mar 16 21:12:08 2009 From: stroucki at ece.cmu.edu (Michael Stroucken) Date: Mon, 16 Mar 2009 21:12:08 -0400 Subject: [Disksim-users] [Fwd: Need help with SSD module] Message-ID: <49BEF8E8.7090007@ece.cmu.edu> Forwarding to disksim-users... -------------- next part -------------- An embedded message was scrubbed... From: "Jin Xin" Subject: Need help with SSD module Date: Tue, 17 Mar 2009 09:53:08 +0900 Size: 8835 URL: From wobber at microsoft.com Mon Mar 16 21:23:27 2009 From: wobber at microsoft.com (Ted Wobber) Date: Mon, 16 Mar 2009 18:23:27 -0700 Subject: [Disksim-users] [Fwd: Need help with SSD module] In-Reply-To: <49BEF8E8.7090007@ece.cmu.edu> References: <49BEF8E8.7090007@ece.cmu.edu> Message-ID: <338B06DAAFBCFD4894EFCAC38E7B811A21F466D71A@NA-EXMSG-C113.redmond.corp.microsoft.com> Unfortunately, the SSD add-on does not include logic to simulate an in-memory cache. I would hope that a cache could be added as a layer on top of the SSD add-on rather than embedding cache code in the SSD module itself. It's even possible that existing disk-cache code could be adapted for this purpose. Best of luck, Ted Wobber MSR Silicon Valley -----Original Message----- From: disksim-users-bounces at ece.cmu.edu [mailto:disksim-users-bounces at ece.cmu.edu] On Behalf Of Michael Stroucken Sent: Monday, March 16, 2009 6:12 PM To: disksim-users at ece.cmu.edu Subject: [Disksim-users] [Fwd: Need help with SSD module] Forwarding to disksim-users... From rimuni at gmail.com Tue Mar 17 16:16:02 2009 From: rimuni at gmail.com (Maliha) Date: Tue, 17 Mar 2009 13:16:02 -0700 Subject: [Disksim-users] Negative transfer time in Seagate Cheetah 15k.5 disk spec Message-ID: <949d418b0903171316y3e7dec94yeb88e84dd729a2dd@mail.gmail.com> Hi I have downloaded the specification of Seagate Cheetah 15k.5 disk from the "on-line database of disk parameters" provided by the disksim team. The disksim_disk block in file Seagate-Cheetah15k5.diskspecs specifies the disk transfer time as follows: Bulk sector transfer time = -0.015000 Does anyone have any idea why the transfer time is negative? (The Bus and Controller transfer times are specified as zero). Thanks in advance, Maliha -------------- next part -------------- An HTML attachment was scrubbed... URL: From janak_ce2005 at yahoo.com Tue Mar 17 21:49:53 2009 From: janak_ce2005 at yahoo.com (janak patel) Date: Tue, 17 Mar 2009 18:49:53 -0700 (PDT) Subject: [Disksim-users] Having problem using TPC-C trace with Disksim Disksim Message-ID: <392094.40756.qm@web110813.mail.gq1.yahoo.com> I would appreciate if someone knows the work around and take some time to answer this question. Thanks! Hi, I am using TPC-C trace but I am not able to make it work. LBA in TPC-C trace is very large than the range of the unsigned long int. How can I make this trace environment suitable to simulation environment? I tried changing the value of 'locScale' in I/O mappings block in parameter file but it doesn't work for me. For example: Request 1: 3247.201 0 276996096 3 0 It gives following error: unexpected request location: devno 0, blkno 276996096, bcount 3 Assertion failed: simtime = 3247.201000 totalreqs = 1 disksim: disksim_logorg.c:763: logorg_maprequest: Assertion `logorgno != -1' failed. Aborted Request 2: 5288.554 0 3201773568 16 0 It gives the following error: unexpected request location: devno 0, blkno 2147483647, bcount 16 Assertion failed: simtime = 5288.554000 totalreqs = 1 disksim: disksim_logorg.c:763: logorg_maprequest: Assertion `logorgno != -1' failed. Aborted In the second request, LBA value gets rounded in the range of long int. How can I fix this problem? Thank you! Janak -------------- next part -------------- An HTML attachment was scrubbed... URL: From luke.chang at intel.com Thu Mar 26 17:08:32 2009 From: luke.chang at intel.com (Chang, Luke) Date: Thu, 26 Mar 2009 14:08:32 -0700 Subject: [Disksim-users] FW: Problems with DiskSim 4.0 Message-ID: <027B85A311E5BD408473843D7A6561571889F31B@orsmsx503.amr.corp.intel.com> Hi: I am working with DiskSim and have run across couple of problems. Are there known fixes to these issues: I am running Debian Linux with g++ version 3.4.3. The first issue I ran into is that during compile, disksim_iosim.c, line 712: StaticAssert (sizeof(ioreq_event) <= DISKSIM_EVENT_SIZE) returns an error: disksim_iosim.c: In function `io_initialize': disksim_iosim.c:712: error: duplicate case value disksim_iosim.c:712: error: previously used here My guess is that the most probable cause for this error is an enumeration problem with ioreq_event in disksim_iosim.h. However, I am not sure what the correct fix is. After working around the first issue by making sizeof(ioreq_event) a constant, then I run into an issue trying to execute runvalid. The disksim.c core dumps and points out the enclosed atlas_III.parv parameter file as invalid format. I checked the format with the user manual and it seemed to have all required elements. The filename also is consistent within disksim_global.h. So why is disksim.c crashing? Thanks Luke -------------- next part -------------- An HTML attachment was scrubbed... URL: From roaslin at gmail.com Mon Mar 30 13:40:09 2009 From: roaslin at gmail.com (Raul) Date: Mon, 30 Mar 2009 19:40:09 +0200 Subject: [Disksim-users] Scheduling Algorithm Message-ID: Hello, I?m testing Disk response Time with cheetah4LP. By default the scheduling algorithm in diskspecs is 1 FCFS, I changed this value to 7 to test SSTF Disk response TIme but the results are the same: 100 consecutive requests , read size = 1blk ===> Disk response Time = 0.224548 100 non consecutive requests, read size = 1blk ===> Disk Response Time = 1.245206 is It correct? must I change de size if request to see changes in times?? thank you From ll2 at rice.edu Mon Mar 30 19:29:29 2009 From: ll2 at rice.edu (Lanyue Lu) Date: Mon, 30 Mar 2009 23:29:29 -0000 Subject: [Disksim-users] Scheduling Algorithm In-Reply-To: References: Message-ID: <4A4A9FBA.5000500@rice.edu> Hi, cheetah4LP disk only has on-disk queue length 1, so no matter what scheduling algorithm used, the performance is the same. You can observe different performance by changing the on-disk maximal queue length in diskspecs file. best, Lanyue Lu Raul wrote: > Hello, > > I?m testing Disk response Time with cheetah4LP. > > By default the scheduling algorithm in diskspecs is 1 FCFS, I changed > this value to 7 to test SSTF Disk response TIme but the results are > the same: > > 100 consecutive requests , read size = 1blk ===> Disk response Time = > 0.224548 > 100 non consecutive requests, read size = 1blk ===> Disk Response Time > = 1.245206 > > > is It correct? > > must I change de size if request to see changes in times?? > > thank you > > _______________________________________________ > Disksim-users mailing list > Disksim-users at ece.cmu.edu > https://sos.ece.cmu.edu/mailman/listinfo/disksim-users >