Now available: The Design and Implementation of the FreeBSD Operating System (Second Edition)
[ source navigation ] [ identifier search ] [ freetext search ] [ file search ] [ list types ] [ track identifier ]
Version: - FREEBSD - FREEBSD-12-STABLE - FREEBSD-12-0 - FREEBSD-11-STABLE - FREEBSD-11-2 - FREEBSD-11-1 - FREEBSD-11-0 - FREEBSD-10-STABLE - FREEBSD-10-4 - FREEBSD-10-3 - FREEBSD-10-2 - FREEBSD-10-1 - FREEBSD-10-0 - FREEBSD-9-STABLE - FREEBSD-9-3 - FREEBSD-9-2 - FREEBSD-9-1 - FREEBSD-9-0 - FREEBSD-8-STABLE - FREEBSD-8-4 - FREEBSD-8-3 - FREEBSD-8-2 - FREEBSD-8-1 - FREEBSD-8-0 - FREEBSD-7-STABLE - FREEBSD-7-4 - FREEBSD-7-3 - FREEBSD-7-2 - FREEBSD-7-1 - FREEBSD-7-0 - FREEBSD-6-STABLE - FREEBSD-6-4 - FREEBSD-6-3 - FREEBSD-6-2 - FREEBSD-6-1 - FREEBSD-6-0 - FREEBSD-5-STABLE - FREEBSD-5-5 - FREEBSD-5-4 - FREEBSD-5-3 - FREEBSD-5-2 - FREEBSD-5-1 - FREEBSD-5-0 - FREEBSD-4-STABLE - FREEBSD-3-STABLE - FREEBSD22 - linux-2.6 - linux-2.4.22 - MK83 - MK84 - PLAN9 - DFBSD - NETBSD - NETBSD5 - NETBSD4 - NETBSD3 - NETBSD20 - OPENBSD - xnu-517 - xnu-792 - xnu-792.6.70 - xnu-1228 - xnu-1456.1.26 - xnu-1699.24.8 - xnu-2050.18.24 - OPENSOLARIS - minix-3-1-1
SearchContext: - none - 3 - 10
1 2 --- GEOM BASED DISK SCHEDULERS FOR FREEBSD --- 3 4 This code contains a framework for GEOM-based disk schedulers and a 5 couple of sample scheduling algorithms that use the framework and 6 implement two forms of "anticipatory scheduling" (see below for more 7 details). 8 9 As a quick example of what this code can give you, try to run "dd", 10 "tar", or some other program with highly SEQUENTIAL access patterns, 11 together with "cvs", "cvsup", "svn" or other highly RANDOM access patterns 12 (this is not a made-up example: it is pretty common for developers 13 to have one or more apps doing random accesses, and others that do 14 sequential accesses e.g., loading large binaries from disk, checking 15 the integrity of tarballs, watching media streams and so on). 16 17 These are the results we get on a local machine (AMD BE2400 dual 18 core CPU, SATA 250GB disk): 19 20 /mnt is a partition mounted on /dev/ad0s1f 21 22 cvs: cvs -d /mnt/home/ncvs-local update -Pd /mnt/ports 23 dd-read: dd bs=128k of=/dev/null if=/dev/ad0 (or ad0-sched-) 24 dd-writew dd bs=128k if=/dev/zero of=/mnt/largefile 25 26 NO SCHEDULER RR SCHEDULER 27 dd cvs dd cvs 28 29 dd-read only 72 MB/s ---- 72 MB/s --- 30 dd-write only 55 MB/s --- 55 MB/s --- 31 dd-read+cvs 6 MB/s ok 30 MB/s ok 32 dd-write+cvs 55 MB/s slooow 14 MB/s ok 33 34 As you can see, when a cvs is running concurrently with dd, the 35 performance drops dramatically, and depending on read or write mode, 36 one of the two is severely penalized. The use of the RR scheduler 37 in this example makes the dd-reader go much faster when competing 38 with cvs, and lets cvs progress when competing with a writer. 39 40 To try it out: 41 42 1. USERS OF FREEBSD 7, PLEASE READ CAREFULLY THE FOLLOWING: 43 44 On loading, this module patches one kernel function (g_io_request()) 45 so that I/O requests ("bio's") carry a classification tag, useful 46 for scheduling purposes. 47 48 ON FREEBSD 7, the tag is stored in an existing (though rarely used) 49 field of the "struct bio", a solution which makes this module 50 incompatible with other modules using it, such as ZFS and gjournal. 51 Additionally, g_io_request() is patched in-memory to add a call 52 to the function that initializes this field (i386/amd64 only; 53 for other architectures you need to manually patch sys/geom/geom_io.c). 54 See details in the file g_sched.c. 55 56 On FreeBSD 8.0 and above, the above trick is not necessary, 57 as the struct bio contains dedicated fields for the classifier, 58 and hooks for request classifiers. 59 60 If you don't like the above, don't run this code. 61 62 2. PLEASE MAKE SURE THAT THE DISK THAT YOU WILL BE USING FOR TESTS 63 DOES NOT CONTAIN PRECIOUS DATA. 64 This is experimental code, so we make no guarantees, though 65 I am routinely using it on my desktop and laptop. 66 67 3. EXTRACT AND BUILD THE PROGRAMS 68 A 'make install' in the directory should work (with root privs), 69 or you can even try the binary modules. 70 If you want to build the modules yourself, look at the Makefile. 71 72 4. LOAD THE MODULE, CREATE A GEOM NODE, RUN TESTS 73 74 The scheduler's module must be loaded first: 75 76 # kldload gsched_rr 77 78 substitute with gsched_as to test AS. Then, supposing that you are 79 using /dev/ad0 for testing, a scheduler can be attached to it with: 80 81 # geom sched insert ad0 82 83 The scheduler is inserted transparently in the geom chain, so 84 mounted partitions and filesystems will keep working, but 85 now requests will go through the scheduler. 86 87 To change scheduler on-the-fly, you can reconfigure the geom: 88 89 # geom sched configure -a as ad0.sched. 90 91 assuming that gsched_as was loaded previously. 92 93 5. SCHEDULER REMOVAL 94 95 In principle it is possible to remove the scheduler module 96 even on an active chain by doing 97 98 # geom sched destroy ad0.sched. 99 100 However, there is some race in the geom subsystem which makes 101 the removal unsafe if there are active requests on a chain. 102 So, in order to reduce the risk of data losses, make sure 103 you don't remove a scheduler from a chain with ongoing transactions. 104 105 --- NOTES ON THE SCHEDULERS --- 106 107 The important contribution of this code is the framework to experiment 108 with different scheduling algorithms. 'Anticipatory scheduling' 109 is a very powerful technique based on the following reasoning: 110 111 The disk throughput is much better if it serves sequential requests. 112 If we have a mix of sequential and random requests, and we see a 113 non-sequential request, do not serve it immediately but instead wait 114 a little bit (2..5ms) to see if there is another one coming that 115 the disk can serve more efficiently. 116 117 There are many details that should be added to make sure that the 118 mechanism is effective with different workloads and systems, to 119 gain a few extra percent in performance, to improve fairness, 120 insulation among processes etc. A discussion of the vast literature 121 on the subject is beyond the purpose of this short note. 122 123 -------------------------------------------------------------------------- 124 125 TRANSPARENT INSERT/DELETE 126 127 geom_sched is an ordinary geom module, however it is convenient 128 to plug it transparently into the geom graph, so that one can 129 enable or disable scheduling on a mounted filesystem, and the 130 names in /etc/fstab do not depend on the presence of the scheduler. 131 132 To understand how this works in practice, remember that in GEOM 133 we have "providers" and "geom" objects. 134 Say that we want to hook a scheduler on provider "ad0", 135 accessible through pointer 'pp'. Originally, pp is attached to 136 geom "ad0" (same name, different object) accessible through pointer old_gp 137 138 BEFORE ---> [ pp --> old_gp ...] 139 140 A normal "geom sched create ad0" call would create a new geom node 141 on top of provider ad0/pp, and export a newly created provider 142 ("ad0.sched." accessible through pointer newpp). 143 144 AFTER create ---> [ newpp --> gp --> cp ] ---> [ pp --> old_gp ... ] 145 146 On top of newpp, a whole tree will be created automatically, and we 147 can e.g. mount partitions on /dev/ad0.sched.s1d, and those requests 148 will go through the scheduler, whereas any partition mounted on 149 the pre-existing device entries will not go through the scheduler. 150 151 With the transparent insert mechanism, the original provider "ad0"/pp 152 is hooked to the newly created geom, as follows: 153 154 AFTER insert ---> [ pp --> gp --> cp ] ---> [ newpp --> old_gp ... ] 155 156 so anything that was previously using provider pp will now have 157 the requests routed through the scheduler node. 158 159 A removal ("geom sched destroy ad0.sched.") will restore the original 160 configuration. 161 162 # $FreeBSD: releng/10.0/sys/geom/sched/README 206497 2010-04-12 16:37:45Z luigi $
This page is part of the FreeBSD/Linux Linux Kernel Cross-Reference, and was automatically generated using a modified version of the LXR engine.