Lines Matching +full:fast +full:- +full:speed
1 zstd(1) -- zstd, zstdmt, unzstd, zstdcat - Compress or decompress .zst files
5 --------
7 `zstd` [*OPTIONS*] [-|_INPUT-FILE_] [-o _OUTPUT-FILE_]
9 `zstdmt` is equivalent to `zstd -T0`
11 `unzstd` is equivalent to `zstd -d`
13 `zstdcat` is equivalent to `zstd -dcf`
17 -----------
18 `zstd` is a fast lossless compression algorithm and data compression tool,
21 `zstd` offers highly configurable compression speed,
22 with fast modes at > 200 MB/s per core,
24 It also features a very fast decoder, with speeds > 500 MB/s per core.
29 - Source files are preserved by default.
30 It's possible to remove them automatically by using the `--rm` command.
31 - When compressing a single file, `zstd` displays progress notifications
33 Use `-q` to turn them off.
34 - `zstd` does not accept input from console,
36 - `zstd` displays a short help page when command line is an error.
37 Use `-q` to turn it off.
41 If no _files_ are given or _file_ is `-`, `zstd` reads from standard input
48 Unless `--stdout` or `-o` is specified, _files_ are written to a new file
61 -------
79 * `-z`, `--compress`:
83 (for example, `unzstd` implies `--decompress`).
84 * `-d`, `--decompress`, `--uncompress`:
86 * `-t`, `--test`:
88 This option is equivalent to `--decompress --stdout` except that the
91 * `-b#`:
93 * `--train FILEs`:
96 * `-l`, `--list`:
99 This command can be augmented with the `-v` modifier.
103 * `-#`:
104 `#` compression level \[1-19] (default: 3)
105 * `--ultra`:
108 * `--fast[=#]`:
109 switch to ultra-fast compression levels.
111 The higher the value, the faster the compression speed,
114 Similarly, if a compression level is set after `--fast`, it overrides it.
115 * `-T#`, `--threads=#`:
119 which is either 64 in 32-bit mode, or 256 for 64-bit environments.
121 * `--single-thread`:
124 (This is different from `-T1`, which spawns 1 compression thread in parallel of I/O).
126 Single-thread mode features lower memory usage.
127 Final compressed result is slightly different from `-T1`.
128 * `--auto-threads={physical,logical} (default: physical)`:
129 When using a default amount of threads via `-T0`, choose the default based on the number
131 * `--adapt[=min=#,max=#]` :
133 Compression level adaptation can be observed live by using command `-v`.
135 The feature works when combined with multi-threading and `--long` mode.
136 It does not work with `--single-thread`.
139 _note_ : at the time of this writing, `--adapt` can remain stuck at low speed
141 * `--long[=#]`:
149 Note: If `windowLog` is set to larger than 27, `--long=windowLog` or
150 `--memory=windowSize` needs to be passed to the decompressor.
151 * `-D DICT`:
153 * `--patch-from FILE`:
158 Note: cannot use both this and -D together
159 Note: `--long` mode will be automatically activated if chainLog < fileLog
162 Node: for all levels, you can use --patch-from in --single-thread mode
163 to improve compression ratio at the cost of speed
165 of speed by specifying `--zstd=targetLength=` to be something large
166 (i.e 4096), and by setting a large `--zstd=chainLog=`
167 * `--rsyncable` :
169 compressed file more rsync-friendly. There is a negligible impact to
171 compression speed hit.
172 This feature does not work with `--single-thread`. You probably don't want
175 * `-C`, `--[no-]check`:
177 * `--[no-]content-size`:
180 --content-size (meaning that the original size will be placed in the header).
181 * `--no-dictID`:
185 * `-M#`, `--memory=#`:
191 This is also used during compression when using with --patch-from=. In this case,
197 * `--stream-size=#` :
202 * `--size-hint=#`:
209 * `-o FILE`:
211 * `-f`, `--force`:
214 * `-c`, `--stdout`:
216 * `--[no-]sparse`:
219 Creating sparse files may save disk space and speed up decompression by
224 * `--rm`:
226 …-o, will trigger a confirmation prompt (which can be silenced with -f), as this is a destructive o…
227 * `-k`, `--keep`:
230 * `-r`:
236 * `--filelist FILE`
239 * `--output-dir-flat DIR`:
245 while in combination with `-f`, the last file will be present instead.
246 * `--output-dir-mirror DIR`:
247 similar to `--output-dir-flat`,
253 it will be stored into the "output-dir/var/tmp/abc".
255 name collision resolution will follow the same rules as `--output-dir-flat`.
256 * `--format=FORMAT`:
261 * `-h`/`-H`, `--help`:
263 * `-V`, `--version`:
265 Advanced : `-vV` also displays supported formats.
266 `-vvV` also displays POSIX support.
267 `-q` will only display the version number, suitable for machine reading.
268 * `-v`, `--verbose`:
270 * `-q`, `--quiet`:
273 * `--no-progress`:
275 * `--show-default-cparams`:
280 * `--`:
281 All arguments after `--` are treated as files
300 `-#` for compression level and `-T#` for number of compression threads.
304 ------------------
310 using command `-D dictionaryFileName`.
313 * `--train FILEs`:
318 `--train` can be combined with `-r` to indicate a directory rather than listing all the files,
321 `--train` supports multithreading if `zstd` is compiled with threading support (default).
322 Additional parameters can be specified with `--train-fastcover`.
323 The legacy dictionary builder can be accessed with `--train-legacy`.
324 The slower cover dictionary builder can be accessed with `--train-cover`.
325 Default is equivalent to `--train-fastcover=d=8,steps=4`.
326 * `-o file`:
328 * `--maxdict=#`:
330 * `-#`:
334 * `-B#`:
336 * `-M#`, `--memory=#`:
338 * `--dictID=#`:
341 By default, zstd will create a 4-bytes random number ID.
348 * `--train-cover[=k#,d=#,steps=#,split=#,shrink[=#]]`:
374 `zstd --train-cover FILEs`
376 `zstd --train-cover=k=50,d=8 FILEs`
378 `zstd --train-cover=d=8,steps=500 FILEs`
380 `zstd --train-cover=k=50 FILEs`
382 `zstd --train-cover=k=50,split=60 FILEs`
384 `zstd --train-cover=shrink FILEs`
386 `zstd --train-cover=shrink=2 FILEs`
388 * `--train-fastcover[=k#,d=#,f=#,steps=#,split=#,accel=#]`:
398 The subsegment is hashed to an index in the range [0,2^_f_ - 1].
404 `zstd --train-fastcover FILEs`
406 `zstd --train-fastcover=d=8,f=15,accel=2 FILEs`
408 * `--train-legacy[=selectivity=#]`:
413 `--train-legacy=s=#` is also accepted.
417 `zstd --train-legacy FILEs`
419 `zstd --train-legacy=selectivity=8 FILEs`
423 ---------
425 * `-b#`:
427 * `-e#`:
428 benchmark file(s) using multiple compression levels, from `-b#` to `-e#` (inclusive)
429 * `-i#`:
431 * `-B#`, `--block-size=#`:
433 * `--priority=rt`:
434 set process priority to real-time
436 **Output Format:** CompressionLevel#Filename : IntputSize -> OutputSize (CompressionRatio), Compres…
438 …or both compression and decompression speed, the entire input is compressed/decompressed in-memory…
441 ----------------------------
442 ### -B#:
444 This parameter is only available when multi-threading is enabled.
447 `-B#` makes it possible to manually select a custom size.
452 ### --zstd[=options]:
456 The _options_ are provided as a comma-separated list.
461 - `strategy`=_strat_, `strat`=_strat_:
469 - `windowLog`=_wlog_, `wlog`=_wlog_:
475 The minimum _wlog_ is 10 (1 KiB) and the maximum is 30 (1 GiB) on 32-bit
476 platforms and 31 (2 GiB) on 64-bit platforms.
478 Note: If `windowLog` is set to larger than 27, `--long=windowLog` or
479 `--memory=windowSize` needs to be passed to the decompressor.
481 - `hashLog`=_hlog_, `hlog`=_hlog_:
489 - `chainLog`=_clog_, `clog`=_clog_:
494 It also slows down compression speed and increases memory requirements for
498 The minimum _clog_ is 6 (64 B) and the maximum is 29 (524 Mib) on 32-bit platforms
499 and 30 (1 Gib) on 64-bit platforms.
501 - `searchLog`=_slog_, `slog`=_slog_:
506 compression ratio but decreases compression speed.
508 The minimum _slog_ is 1 and the maximum is 'windowLog' - 1.
510 - `minMatch`=_mml_, `mml`=_mml_:
514 decompression speed.
518 - `targetLength`=_tlen_, `tlen`=_tlen_:
524 but decreases compression speed.
526 For ZSTD\_fast, it triggers ultra-fast mode when > 0.
528 Impact is reversed : a larger `targetLength` increases compression speed
535 - `overlapLog`=_ovlog_, `ovlog`=_ovlog_:
538 Reloading more data improves compression ratio, but decreases speed.
548 - `ldmHashLog`=_lhlog_, `lhlog`=_lhlog_:
554 memory during compression and a decrease in compression speed.
558 - `ldmMinMatch`=_lmml_, `lmml`=_lmml_:
567 - `ldmBucketSizeLog`=_lblog_, `lblog`=_lblog_:
574 speed.
578 - `ldmHashRateLog`=_lhrlog_, `lhrlog`=_lhrlog_:
584 Larger values will improve compression speed. Deviating far from the
587 The default value is `wlog - lhlog`.
593 `--zstd`=wlog=23,clog=23,hlog=22,slog=6,mml=3,tlen=48,strat=6
597 ----
601 ------