mirrors/qemu
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
34a68001f1
qemu/migration
History
Andrei Gudkov 34a68001f1 migration/calc-dirty-rate: millisecond-granularity period 
This patch allows to measure dirty page rate for
sub-second intervals of time. An optional argument is
introduced -- calc-time-unit. For example:
{"execute": "calc-dirty-rate", "arguments":
  {"calc-time": 500, "calc-time-unit": "millisecond"} }

Millisecond granularity allows to make predictions whether
migration will succeed or not. To do this, calculate dirty
rate with calc-time set to max allowed downtime (e.g. 300ms),
convert measured rate into volume of dirtied memory,
and divide by network throughput. If the value is lower
than max allowed downtime, then migration will converge.

Measurement results for single thread randomly writing to
a 1/4/24GiB memory region:

+----------------+-----------------------------------------------+
| calc-time      |                dirty rate MiB/s               |
| (milliseconds) +----------------+---------------+--------------+
|                | theoretical    | page-sampling | dirty-bitmap |
|                | (at 3M wr/sec) |               |              |
+----------------+----------------+---------------+--------------+
|                               1GiB                             |
+----------------+----------------+---------------+--------------+
|            100 |           6996 |          7100 |         3192 |
|            200 |           4606 |          4660 |         2655 |
|            300 |           3305 |          3280 |         2371 |
|            400 |           2534 |          2525 |         2154 |
|            500 |           2041 |          2044 |         1871 |
|            750 |           1365 |          1341 |         1358 |
|           1000 |           1024 |          1052 |         1025 |
|           1500 |            683 |           678 |          684 |
|           2000 |            512 |           507 |          513 |
+----------------+----------------+---------------+--------------+
|                               4GiB                             |
+----------------+----------------+---------------+--------------+
|            100 |          10232 |          8880 |         4070 |
|            200 |           8954 |          8049 |         3195 |
|            300 |           7889 |          7193 |         2881 |
|            400 |           6996 |          6530 |         2700 |
|            500 |           6245 |          5772 |         2312 |
|            750 |           4829 |          4586 |         2465 |
|           1000 |           3865 |          3780 |         2178 |
|           1500 |           2694 |          2633 |         2004 |
|           2000 |           2041 |          2031 |         1789 |
+----------------+----------------+---------------+--------------+
|                               24GiB                            |
+----------------+----------------+---------------+--------------+
|            100 |          11495 |          8640 |         5597 |
|            200 |          11226 |          8616 |         3527 |
|            300 |          10965 |          8386 |         2355 |
|            400 |          10713 |          8370 |         2179 |
|            500 |          10469 |          8196 |         2098 |
|            750 |           9890 |          7885 |         2556 |
|           1000 |           9354 |          7506 |         2084 |
|           1500 |           8397 |          6944 |         2075 |
|           2000 |           7574 |          6402 |         2062 |
+----------------+----------------+---------------+--------------+

Theoretical values are computed according to the following formula:
size * (1 - (1-(4096/size))^(time*wps)) / (time * 2^20),
where size is in bytes, time is in seconds, and wps is number of
writes per second.

Signed-off-by: Andrei Gudkov <gudkov.andrei@huawei.com>
Reviewed-by: Hyman Huang <yong.huang@smartx.com>
Message-Id: <d802e6b8053eb60fbec1a784cf86f67d9528e0a8.1693895970.git.gudkov.andrei@huawei.com>
Signed-off-by: Hyman Huang <yong.huang@smartx.com>
2023-10-10 08:03:50 +08:00
..
block-dirty-bitmap.c migration: Move rate_limit_max and rate_limit_used to migration_stats 2023-05-18 18:40:51 +02:00
block.c Migration Pull request (20231002) 2023-10-02 14:42:44 -04:00
block.h migration: disable auto-converge during bulk block migration 2017-09-27 11:27:14 +01:00
channel-block.c io: follow coroutine AioContext in qio_channel_yield() 2023-09-07 20:32:11 -05:00
channel-block.h migration: introduce a QIOChannel impl for BlockDriverState VMState 2022-06-22 19:33:43 +01:00
channel.c migration: check magic value for deciding the mapping of channels 2023-02-06 19:22:57 +01:00
channel.h migration: check magic value for deciding the mapping of channels 2023-02-06 19:22:57 +01:00
colo-failover.c migration/colo: Improve an x-colo-lost-heartbeat error message 2023-02-23 14:10:17 +01:00
colo.c migration: process_incoming_migration_co(): move colo part to colo 2023-05-18 18:40:51 +02:00
dirtyrate.c migration/calc-dirty-rate: millisecond-granularity period 2023-10-10 08:03:50 +08:00
dirtyrate.h migration/calc-dirty-rate: millisecond-granularity period 2023-10-10 08:03:50 +08:00
exec.c *: Add missing includes of qemu/error-report.h 2023-03-22 15:06:57 +00:00
exec.h migration: Export exec.c functions in its own file 2017-06-01 18:49:22 +02:00
fd.c bulk: Remove pointless QOM casts 2023-06-05 20:48:34 +02:00
fd.h migration: Fix fd protocol for incoming defer 2019-06-05 12:43:55 +02:00
file.c migration: file URI offset 2023-10-04 13:18:08 +02:00
file.h migration: file URI 2023-10-04 13:16:58 +02:00
global_state.c migration: never fail in global_state_store() 2023-06-02 01:03:19 +02:00
meson.build migration: file URI 2023-10-04 13:16:58 +02:00
migration-hmp-cmds.c migration: Extend query-migrate to provide dirty page limit info 2023-07-26 10:55:56 +02:00
migration-stats.c migration: Don't abuse qemu_file transferred for RDMA 2023-09-29 18:11:21 +02:00
migration-stats.h migration: Don't abuse qemu_file transferred for RDMA 2023-09-29 18:11:21 +02:00
migration.c migration: file URI 2023-10-04 13:16:58 +02:00
migration.h migration: Replace the return path retry logic 2023-09-27 13:58:02 -04:00
multifd-zlib.c migration: spelling fixes 2023-07-25 17:13:20 +03:00
multifd-zstd.c migration: spelling fixes 2023-07-25 17:13:20 +03:00
multifd.c migration/multifd: Rename threadinfo.c functions 2023-07-26 10:55:56 +02:00
multifd.h multifd: Add the ramblock to MultiFDRecvParams 2023-05-10 18:48:11 +02:00
options.c migration: enforce multifd and postcopy preempt to be set before incoming 2023-07-26 10:55:56 +02:00
options.h migration: Introduce dirty-limit capability 2023-07-26 10:55:56 +02:00
page_cache.c migration: Fix cache_init()'s "Failed to allocate" error messages 2021-02-08 11:19:51 +00:00
page_cache.h migration: Clean up signed vs. unsigned XBZRLE cache-size 2021-02-08 11:19:51 +00:00
postcopy-ram.c migration: Fix race that dest preempt thread close too early 2023-09-27 13:58:02 -04:00
postcopy-ram.h migration: Allow postcopy_ram_supported_by_host() to report err 2023-04-27 10:18:25 +02:00
qemu-file.c migration/rdma: Simplify the function that saves a page 2023-09-29 18:13:53 +02:00
qemu-file.h migration/rdma: Simplify the function that saves a page 2023-09-29 18:13:53 +02:00
ram-compress.c ram-compress.c: Make target independent 2023-05-08 15:25:26 +02:00
ram-compress.h ram.c: Move core decompression code into its own file 2023-05-08 15:25:26 +02:00
ram.c Migration Pull request (20231002) 2023-10-02 14:42:44 -04:00
ram.h migration/ram: Expose ramblock_is_ignored() as migrate_ram_is_ignored() 2023-07-12 09:25:37 +02:00
rdma.c migration/rdma: zore out head.repeat to make the error more clear 2023-10-04 10:54:41 +02:00
rdma.h migration: Export rdma.c functions in its own file 2017-06-01 18:49:23 +02:00
savevm.c migration: Update error description outside migration.c 2023-10-04 10:54:40 +02:00
savevm.h migration: Add .save_prepare() handler to struct SaveVMHandlers 2023-09-11 08:34:06 +02:00
socket.c migration: Move migrate_use_zero_copy_send() to options.c 2023-04-24 15:01:46 +02:00
socket.h migration: Postcopy preemption preparation on channel creation 2022-07-20 12:15:08 +01:00
target.c migration: Add migration prefix to functions in target.c 2023-09-11 08:34:06 +02:00
threadinfo.c migration/multifd: Protect accesses to migration_threads 2023-07-26 10:55:56 +02:00
threadinfo.h migration/multifd: Protect accesses to migration_threads 2023-07-26 10:55:56 +02:00
tls.c migration: Drop unused parameter for migration_tls_client_create() 2023-05-03 11:24:20 +02:00
tls.h migration: Drop unused parameter for migration_tls_client_create() 2023-05-03 11:24:20 +02:00
trace-events migration: Unify and trace vmstate field_exists() checks 2023-10-04 13:19:47 +02:00
trace.h trace: switch position of headers to what Meson requires 2020-08-21 06:18:24 -04:00
vmstate-types.c Move CPU softfloat unions to cpu-float.h 2022-04-06 14:31:43 +02:00
vmstate.c migration: Unify and trace vmstate field_exists() checks 2023-10-04 13:19:47 +02:00
xbzrle.c migration/xbzrle: Use i386 host/cpuinfo.h 2023-05-23 16:51:18 -07:00
xbzrle.h migration/xbzrle: Use i386 host/cpuinfo.h 2023-05-23 16:51:18 -07:00
yank_functions.c bulk: Remove pointless QOM casts 2023-06-05 20:48:34 +02:00
yank_functions.h migration: Move the yank unregister of channel_close out 2021-07-26 12:45:03 +01:00