Mercurial > hg
view mercurial/setdiscovery.py @ 42044:bb271ec2fbfb
compression: introduce a `storage.revlog.zstd.level` configuration
This option control the zstd compression level used when compressing revlog
chunk. The usage of zstd for revlog compression has not graduated from
experimental yet, but we intend to fix that soon.
The option name for the compression level is more straight forward to pick, so
this changesets comes first. Having a dedicated option for each compression
engine is useful because they don't support the same range of values.
I ran the same measurement as for the zlib compression level (in the parent
changesets). The variation in repository size is stay mostly in the same (small)
range. The "read/write" performance see smallish variation, but are overall much
better than zlib. Write performance show the same tend of having better write
performance for when reaching high-end compression.
Again, we don't intend to change the default zstd compression level (currently:
3) in this series. However this is worth investigating in the future.
The Performance comparison of zlib vs zstd is quite impressive. The repository
size stay in the same range, but the performance are much better in all
situations.
Comparison summary
==================
We are looking at:
- performance range for zlib
- performance range for zstd
- comparison of default zstd (level-3) to default zlib (level 6)
- comparison of the slowest zstd time to the fastest zlib time
Read performance:
-----------------
| zlib | zstd | cmp | f2s
mercurial | 0.170159 - 0.189219 | 0.144127 - 0.149624 | 80% | 88%
pypy | 2.679217 - 2.768691 | 1.532317 - 1.705044 | 60% | 63%
netbeans | 122.477027 - 141.620281 | 72.996346 - 89.731560 | 58% | 73%
mozilla | 147.867662 - 170.572118 | 91.700995 - 105.853099 | 56% | 71%
Write performance:
------------------
| zlib | zstd | cmp | f2s
mercurial | 53.250304 - 56.2936129 | 40.877025 - 45.677286 | 75% | 86%
pypy | 460.721984 - 476.589918 | 270.545409 - 301.002219 | 63% | 65%
netbeans | 520.560316 - 715.930400 | 370.356311 - 428.329652 | 55% | 82%
mozilla | 739.803002 - 987.056093 | 505.152906 - 591.930683 | 57% | 80%
Raw data
--------
repo alg lvl .hg/store size 00manifest.d read write
mercurial zlib 1 49,402,813 5,963,475 0.170159 53.250304
mercurial zlib 6 47,197,397 5,875,730 0.182820 56.264320
mercurial zlib 9 47,121,596 5,849,781 0.189219 56.293612
mercurial zstd 1 49,737,084 5,966,355 0.144127 40.877025
mercurial zstd 3 48,961,867 5,895,208 0.146376 42.268142
mercurial zstd 5 48,200,592 5,938,676 0.149624 43.162875
mercurial zstd 10 47,833,520 5,913,353 0.145185 44.012489
mercurial zstd 15 47,314,604 5,728,679 0.147686 45.677286
mercurial zstd 20 47,330,502 5,830,539 0.145789 45.025407
mercurial zstd 22 47,330,076 5,830,539 0.143996 44.690460
pypy zlib 1 370,830,572 28,462,425 2.679217 460.721984
pypy zlib 6 340,112,317 27,648,747 2.768691 467.537158
pypy zlib 9 338,360,736 27,639,003 2.763495 476.589918
pypy zstd 1 362,377,479 27,916,214 1.532317 270.545409
pypy zstd 3 354,137,693 27,905,988 1.686718 294.951509
pypy zstd 5 342,640,043 27,655,774 1.705044 301.002219
pypy zstd 10 334,224,327 27,164,493 1.567287 285.186239
pypy zstd 15 329,000,363 26,645,965 1.637729 299.561332
pypy zstd 20 324,534,039 26,199,547 1.526813 302.149827
pypy zstd 22 324,530,595 26,198,932 1.525718 307.821218
netbeans zlib 1 1,281,847,810 165,495,457 122.477027 520.560316
netbeans zlib 6 1,205,284,353 159,161,207 139.876147 715.930400
netbeans zlib 9 1,197,135,671 155,034,586 141.620281 678.297064
netbeans zstd 1 1,259,581,737 160,840,613 72.996346 370.356311
netbeans zstd 3 1,232,978,122 157,691,551 81.622317 396.733087
netbeans zstd 5 1,208,034,075 160,246,880 83.080549 364.342626
netbeans zstd 10 1,188,624,176 156,083,417 79.323935 403.594602
netbeans zstd 15 1,176,973,589 153,859,477 89.731560 428.329652
netbeans zstd 20 1,162,958,258 151,147,535 82.842667 392.335349
netbeans zstd 22 1,162,707,029 151,150,220 82.565695 402.840655
mozilla zlib 1 2,775,497,186 298,527,987 147.867662 751.263721
mozilla zlib 6 2,596,856,420 286,597,671 170.572118 987.056093
mozilla zlib 9 2,587,542,494 287,018,264 163.622338 739.803002
mozilla zstd 1 2,723,159,348 286,617,532 91.700995 570.042751
mozilla zstd 3 2,665,055,001 286,152,013 95.240155 561.412805
mozilla zstd 5 2,607,819,817 288,060,030 101.978048 505.152906
mozilla zstd 10 2,558,761,085 283,967,648 104.113481 497.771202
mozilla zstd 15 2,526,216,060 275,581,300 105.853099 591.930683
mozilla zstd 20 2,485,114,806 266,478,859 95.268795 576.515389
mozilla zstd 22 2,484,869,080 266,456,505 94.429282 572.785537
author | Pierre-Yves David <pierre-yves.david@octobus.net> |
---|---|
date | Wed, 27 Mar 2019 18:35:59 +0100 |
parents | 0d467e4de4ae |
children | 362726923ba3 |
line wrap: on
line source
# setdiscovery.py - improved discovery of common nodeset for mercurial # # Copyright 2010 Benoit Boissinot <bboissin@gmail.com> # and Peter Arrenbrecht <peter@arrenbrecht.ch> # # This software may be used and distributed according to the terms of the # GNU General Public License version 2 or any later version. """ Algorithm works in the following way. You have two repository: local and remote. They both contains a DAG of changelists. The goal of the discovery protocol is to find one set of node *common*, the set of nodes shared by local and remote. One of the issue with the original protocol was latency, it could potentially require lots of roundtrips to discover that the local repo was a subset of remote (which is a very common case, you usually have few changes compared to upstream, while upstream probably had lots of development). The new protocol only requires one interface for the remote repo: `known()`, which given a set of changelists tells you if they are present in the DAG. The algorithm then works as follow: - We will be using three sets, `common`, `missing`, `unknown`. Originally all nodes are in `unknown`. - Take a sample from `unknown`, call `remote.known(sample)` - For each node that remote knows, move it and all its ancestors to `common` - For each node that remote doesn't know, move it and all its descendants to `missing` - Iterate until `unknown` is empty There are a couple optimizations, first is instead of starting with a random sample of missing, start by sending all heads, in the case where the local repo is a subset, you computed the answer in one round trip. Then you can do something similar to the bisecting strategy used when finding faulty changesets. Instead of random samples, you can try picking nodes that will maximize the number of nodes that will be classified with it (since all ancestors or descendants will be marked as well). """ from __future__ import absolute_import import collections import random from .i18n import _ from .node import ( nullid, nullrev, ) from . import ( error, util, ) def _updatesample(revs, heads, sample, parentfn, quicksamplesize=0): """update an existing sample to match the expected size The sample is updated with revs exponentially distant from each head of the <revs> set. (H~1, H~2, H~4, H~8, etc). If a target size is specified, the sampling will stop once this size is reached. Otherwise sampling will happen until roots of the <revs> set are reached. :revs: set of revs we want to discover (if None, assume the whole dag) :heads: set of DAG head revs :sample: a sample to update :parentfn: a callable to resolve parents for a revision :quicksamplesize: optional target size of the sample""" dist = {} visit = collections.deque(heads) seen = set() factor = 1 while visit: curr = visit.popleft() if curr in seen: continue d = dist.setdefault(curr, 1) if d > factor: factor *= 2 if d == factor: sample.add(curr) if quicksamplesize and (len(sample) >= quicksamplesize): return seen.add(curr) for p in parentfn(curr): if p != nullrev and (not revs or p in revs): dist.setdefault(p, d + 1) visit.append(p) def _limitsample(sample, desiredlen): """return a random subset of sample of at most desiredlen item""" if len(sample) > desiredlen: sample = set(random.sample(sample, desiredlen)) return sample class partialdiscovery(object): """an object representing ongoing discovery Feed with data from the remote repository, this object keep track of the current set of changeset in various states: - common: revs also known remotely - undecided: revs we don't have information on yet - missing: revs missing remotely (all tracked revisions are known locally) """ def __init__(self, repo, targetheads): self._repo = repo self._targetheads = targetheads self._common = repo.changelog.incrementalmissingrevs() self._undecided = None self.missing = set() self._childrenmap = None def addcommons(self, commons): """registrer nodes known as common""" self._common.addbases(commons) if self._undecided is not None: self._common.removeancestorsfrom(self._undecided) def addmissings(self, missings): """registrer some nodes as missing""" newmissing = self._repo.revs('%ld::%ld', missings, self.undecided) if newmissing: self.missing.update(newmissing) self.undecided.difference_update(newmissing) def addinfo(self, sample): """consume an iterable of (rev, known) tuples""" common = set() missing = set() for rev, known in sample: if known: common.add(rev) else: missing.add(rev) if common: self.addcommons(common) if missing: self.addmissings(missing) def hasinfo(self): """return True is we have any clue about the remote state""" return self._common.hasbases() def iscomplete(self): """True if all the necessary data have been gathered""" return self._undecided is not None and not self._undecided @property def undecided(self): if self._undecided is not None: return self._undecided self._undecided = set(self._common.missingancestors(self._targetheads)) return self._undecided def commonheads(self): """the heads of the known common set""" # heads(common) == heads(common.bases) since common represents # common.bases and all its ancestors return self._common.basesheads() def _parentsgetter(self): getrev = self._repo.changelog.index.__getitem__ def getparents(r): return getrev(r)[5:7] return getparents def _childrengetter(self): if self._childrenmap is not None: # During discovery, the `undecided` set keep shrinking. # Therefore, the map computed for an iteration N will be # valid for iteration N+1. Instead of computing the same # data over and over we cached it the first time. return self._childrenmap.__getitem__ # _updatesample() essentially does interaction over revisions to look # up their children. This lookup is expensive and doing it in a loop is # quadratic. We precompute the children for all relevant revisions and # make the lookup in _updatesample() a simple dict lookup. self._childrenmap = children = {} parentrevs = self._parentsgetter() revs = self.undecided for rev in sorted(revs): # Always ensure revision has an entry so we don't need to worry # about missing keys. children[rev] = [] for prev in parentrevs(rev): if prev == nullrev: continue c = children.get(prev) if c is not None: c.append(rev) return children.__getitem__ def takequicksample(self, headrevs, size): """takes a quick sample of size <size> It is meant for initial sampling and focuses on querying heads and close ancestors of heads. :headrevs: set of head revisions in local DAG to consider :size: the maximum size of the sample""" revs = self.undecided if len(revs) <= size: return list(revs) sample = set(self._repo.revs('heads(%ld)', revs)) if len(sample) >= size: return _limitsample(sample, size) _updatesample(None, headrevs, sample, self._parentsgetter(), quicksamplesize=size) return sample def takefullsample(self, headrevs, size): revs = self.undecided if len(revs) <= size: return list(revs) repo = self._repo sample = set(repo.revs('heads(%ld)', revs)) parentrevs = self._parentsgetter() # update from heads revsheads = sample.copy() _updatesample(revs, revsheads, sample, parentrevs) # update from roots revsroots = set(repo.revs('roots(%ld)', revs)) childrenrevs = self._childrengetter() _updatesample(revs, revsroots, sample, childrenrevs) assert sample sample = _limitsample(sample, size) if len(sample) < size: more = size - len(sample) sample.update(random.sample(list(revs - sample), more)) return sample def findcommonheads(ui, local, remote, initialsamplesize=100, fullsamplesize=200, abortwhenunrelated=True, ancestorsof=None): '''Return a tuple (common, anyincoming, remoteheads) used to identify missing nodes from or in remote. ''' start = util.timer() roundtrips = 0 cl = local.changelog clnode = cl.node clrev = cl.rev if ancestorsof is not None: ownheads = [clrev(n) for n in ancestorsof] else: ownheads = [rev for rev in cl.headrevs() if rev != nullrev] # early exit if we know all the specified remote heads already ui.debug("query 1; heads\n") roundtrips += 1 sample = _limitsample(ownheads, initialsamplesize) # indices between sample and externalized version must match sample = list(sample) with remote.commandexecutor() as e: fheads = e.callcommand('heads', {}) fknown = e.callcommand('known', { 'nodes': [clnode(r) for r in sample], }) srvheadhashes, yesno = fheads.result(), fknown.result() if cl.tip() == nullid: if srvheadhashes != [nullid]: return [nullid], True, srvheadhashes return [nullid], False, [] # start actual discovery (we note this before the next "if" for # compatibility reasons) ui.status(_("searching for changes\n")) knownsrvheads = [] # revnos of remote heads that are known locally for node in srvheadhashes: if node == nullid: continue try: knownsrvheads.append(clrev(node)) # Catches unknown and filtered nodes. except error.LookupError: continue if len(knownsrvheads) == len(srvheadhashes): ui.debug("all remote heads known locally\n") return srvheadhashes, False, srvheadhashes if len(sample) == len(ownheads) and all(yesno): ui.note(_("all local heads known remotely\n")) ownheadhashes = [clnode(r) for r in ownheads] return ownheadhashes, True, srvheadhashes # full blown discovery disco = partialdiscovery(local, ownheads) # treat remote heads (and maybe own heads) as a first implicit sample # response disco.addcommons(knownsrvheads) disco.addinfo(zip(sample, yesno)) full = False progress = ui.makeprogress(_('searching'), unit=_('queries')) while not disco.iscomplete(): if full or disco.hasinfo(): if full: ui.note(_("sampling from both directions\n")) else: ui.debug("taking initial sample\n") samplefunc = disco.takefullsample targetsize = fullsamplesize else: # use even cheaper initial sample ui.debug("taking quick initial sample\n") samplefunc = disco.takequicksample targetsize = initialsamplesize sample = samplefunc(ownheads, targetsize) roundtrips += 1 progress.update(roundtrips) ui.debug("query %i; still undecided: %i, sample size is: %i\n" % (roundtrips, len(disco.undecided), len(sample))) # indices between sample and externalized version must match sample = list(sample) with remote.commandexecutor() as e: yesno = e.callcommand('known', { 'nodes': [clnode(r) for r in sample], }).result() full = True disco.addinfo(zip(sample, yesno)) result = disco.commonheads() elapsed = util.timer() - start progress.complete() ui.debug("%d total queries in %.4fs\n" % (roundtrips, elapsed)) msg = ('found %d common and %d unknown server heads,' ' %d roundtrips in %.4fs\n') missing = set(result) - set(knownsrvheads) ui.log('discovery', msg, len(result), len(missing), roundtrips, elapsed) if not result and srvheadhashes != [nullid]: if abortwhenunrelated: raise error.Abort(_("repository is unrelated")) else: ui.warn(_("warning: repository is unrelated\n")) return ({nullid}, True, srvheadhashes,) anyincoming = (srvheadhashes != [nullid]) result = {clnode(r) for r in result} return result, anyincoming, srvheadhashes