You can not select more than 25 topics
Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
442 lines
20 KiB
442 lines
20 KiB
// PeerSelection.java
|
|
// -------------------------------------
|
|
// (C) by Michael Peter Christen; mc@yacy.net
|
|
// first published 05.11.2008 on http://yacy.net
|
|
// Frankfurt, Germany, 2008
|
|
//
|
|
// $LastChangedDate$
|
|
// $LastChangedRevision$
|
|
// $LastChangedBy$
|
|
//
|
|
// This program is free software; you can redistribute it and/or modify
|
|
// it under the terms of the GNU General Public License as published by
|
|
// the Free Software Foundation; either version 2 of the License, or
|
|
// (at your option) any later version.
|
|
//
|
|
// This program is distributed in the hope that it will be useful,
|
|
// but WITHOUT ANY WARRANTY; without even the implied warranty of
|
|
// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
|
// GNU General Public License for more details.
|
|
//
|
|
// You should have received a copy of the GNU General Public License
|
|
// along with this program; if not, write to the Free Software
|
|
// Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
|
|
|
|
package net.yacy.peers;
|
|
|
|
import java.lang.reflect.Array;
|
|
import java.util.ArrayList;
|
|
import java.util.Collection;
|
|
import java.util.HashSet;
|
|
import java.util.Iterator;
|
|
import java.util.LinkedHashSet;
|
|
import java.util.List;
|
|
import java.util.Random;
|
|
import java.util.Set;
|
|
import java.util.SortedSet;
|
|
import java.util.concurrent.ConcurrentHashMap;
|
|
import java.util.concurrent.ConcurrentMap;
|
|
|
|
import net.yacy.cora.document.encoding.ASCII;
|
|
import net.yacy.cora.federate.yacy.Distribution;
|
|
import net.yacy.cora.order.Base64Order;
|
|
import net.yacy.cora.order.Digest;
|
|
import net.yacy.cora.sorting.OrderedScoreMap;
|
|
import net.yacy.cora.storage.HandleSet;
|
|
import net.yacy.cora.util.LookAheadIterator;
|
|
import net.yacy.kelondro.data.word.Word;
|
|
import net.yacy.kelondro.util.kelondroException;
|
|
import net.yacy.peers.operation.yacyVersion;
|
|
|
|
/**
|
|
* This package is a collection of peer selection iterations that had been
|
|
* part of yacyPeerActions, yacyDHTActions and yacySeedDB
|
|
*/
|
|
public class DHTSelection {
|
|
|
|
public static Set<Seed> selectClusterPeers(final SeedDB seedDB, final SortedSet<byte[]> peerhashes) {
|
|
final Set<Seed> l = new HashSet<Seed>();
|
|
Seed s;
|
|
for (byte[] hashb: peerhashes) {
|
|
s = seedDB.get(ASCII.String(hashb)); // should be getConnected; get only during testing time
|
|
if (s != null) l.add(s);
|
|
}
|
|
return l;
|
|
}
|
|
|
|
/**
|
|
*
|
|
* @param seedDB
|
|
* @param wordhashes
|
|
* @param minage
|
|
* @param omit
|
|
* @param maxcount
|
|
* @param r we must use a random factor for the selection to prevent that all peers do the same and therefore overload the same peers
|
|
* @return
|
|
*/
|
|
public static Collection<Seed> selectExtraTargets(
|
|
final SeedDB seedDB,
|
|
final HandleSet wordhashes,
|
|
final int minage,
|
|
final Collection<Seed> omit,
|
|
final int maxcount,
|
|
final Random r) {
|
|
|
|
Collection<Seed> extraSeeds = new HashSet<Seed>();
|
|
|
|
if (seedDB != null) {
|
|
final OrderedScoreMap<Seed> seedSelection = new OrderedScoreMap<Seed>(null);
|
|
|
|
// create sets that contains only robinson/node/large/young peers
|
|
Iterator<Seed> dhtEnum = seedDB.seedsConnected(true, false, null, 0.50f);
|
|
Seed seed;
|
|
while (dhtEnum.hasNext()) {
|
|
seed = dhtEnum.next();
|
|
if (seed == null) continue;
|
|
if (omit != null && omit.contains(seed)) continue; // sort out peers that are target for DHT
|
|
if (seed.isLastSeenTimeout(3600000)) continue; // do not ask peers that had not been seen more than one hour (happens during a startup situation)
|
|
if (!seed.getFlagSolrAvailable()) continue; // extra peers always use solr direct, skip if solr interface is not available
|
|
if (!seed.getFlagAcceptRemoteIndex() && seed.matchPeerTags(wordhashes)) seedSelection.dec(seed, r.nextInt(10) + 2); // robinson peers with matching peer tags
|
|
if (seed.getFlagRootNode()) seedSelection.dec(seed, r.nextInt(30) + 6); // root nodes (fast peers)
|
|
if (seed.getAge() < minage) seedSelection.dec(seed, r.nextInt(15) + 3); // young peers (with fresh info)
|
|
if (seed.getAge() < 1) seedSelection.dec(seed, r.nextInt(40) + 8); // the 'workshop feature', fresh peers should be seen
|
|
if (seed.getLinkCount() >= 100000 && seed.getLinkCount() < 1000000) { // peers above 100.000 links take part on a selection of medium-size peers
|
|
seedSelection.dec(seed, r.nextInt(25) + 5);
|
|
}
|
|
if (seed.getLinkCount() >= 1000000) { // peers above 1 million links take part on a selection of large peers
|
|
int pf = 1 + (int) (20000000 / seed.getLinkCount());
|
|
seedSelection.dec(seed, r.nextInt(pf) + pf / 5); // large peers; choose large one less frequent to reduce load on their peer
|
|
}
|
|
}
|
|
|
|
// select the maxount
|
|
Iterator<Seed> i = seedSelection.iterator();
|
|
int count = 0;
|
|
while (i.hasNext() && count++ < maxcount) {
|
|
seed = i.next();
|
|
if (RemoteSearch.log.isInfo()) {
|
|
RemoteSearch.log.info("selectPeers/extra: " + seed.hash + ":" + seed.getName() + ", " + seed.getLinkCount() + " URLs" +
|
|
(seed.getLinkCount() >= 1000000 ? " LARGE-SIZE" : "") +
|
|
(seed.getLinkCount() >= 100000 && seed.getLinkCount() < 1000000 ? " MEDIUM-SIZE" : "") +
|
|
(!seed.getFlagAcceptRemoteIndex() && seed.matchPeerTags(wordhashes) ? " ROBINSON" : "") +
|
|
(seed.getFlagRootNode() ? " NODE" : "") +
|
|
(seed.getAge() < 1 ? " FRESH" : "")
|
|
);
|
|
}
|
|
extraSeeds.add(seed);
|
|
}
|
|
}
|
|
|
|
return extraSeeds;
|
|
}
|
|
|
|
/**
|
|
* @param seedDB the seeds database.
|
|
* @param wordhashes hashes of the words we are searching for
|
|
* @param minage the minimum age of each seed in days
|
|
* @param minWordCount the minimum RWI words count of each seed
|
|
* @param random a random generator instance
|
|
* @return a list of matching candidate seeds for remote RWI search
|
|
*/
|
|
public static Set<Seed> selectDHTSearchTargets(final SeedDB seedDB, final HandleSet wordhashes, final int minage, final int minWordCount, final int redundancy, final int maxredundancy, final Random random) {
|
|
|
|
// put in seeds according to dht
|
|
Set<Seed> seeds = new LinkedHashSet<>(); // dht position seeds
|
|
if (seedDB != null) {
|
|
Iterator<byte[]> iter = wordhashes.iterator();
|
|
while (iter.hasNext()) {
|
|
seeds.addAll(collectHorizontalDHTPositions(seedDB, iter.next(), minage, minWordCount, redundancy, maxredundancy, random));
|
|
}
|
|
}
|
|
|
|
return seeds;
|
|
}
|
|
|
|
private static ArrayList<Seed> collectHorizontalDHTPositions(final SeedDB seedDB, final byte[] wordhash, final int minage, final int minWordCount, final int redundancy, final int maxredundancy, final Random random) {
|
|
// this method is called from the search target computation
|
|
ArrayList<Seed> collectedSeeds = new ArrayList<>(redundancy * seedDB.scheme.verticalPartitions());
|
|
for (int verticalPosition = 0; verticalPosition < seedDB.scheme.verticalPartitions(); verticalPosition++) {
|
|
ArrayList<Seed> seeds = selectVerticalDHTPositions(seedDB, wordhash, minage, minWordCount, maxredundancy, verticalPosition);
|
|
if (seeds.size() <= redundancy) {
|
|
collectedSeeds.addAll(seeds);
|
|
} else {
|
|
// we pick some random peers from the vertical position.
|
|
// All of them should be valid, but picking a random subset is a distributed load balancing on the whole YaCy network.
|
|
// without picking a random subset, always the same peers would be targeted for the same word resulting in (possible) DoS on the target.
|
|
for (int i = 0; i < redundancy; i++) {
|
|
collectedSeeds.add(seeds.remove(random.nextInt(seeds.size())));
|
|
}
|
|
}
|
|
}
|
|
return collectedSeeds;
|
|
}
|
|
|
|
/**
|
|
* @param seedDB the seeds database. Must not be null.
|
|
* @param wordhash the word we are searching for
|
|
* @param minage the minimum age of each seed in days
|
|
* @param redundancy the number of redundant peer position for this parition, minimum is 1
|
|
* @return a list of matching candidate seeds for DHT distribution
|
|
*/
|
|
@SuppressWarnings("unchecked")
|
|
public static List<Seed>[] selectDHTDistributionTargets(final SeedDB seedDB, final byte[] wordhash, final int minage, final int redundancy) {
|
|
// this method is called from the distribution target computation
|
|
List<Seed>[] seedlists = (List<Seed>[]) Array.newInstance(ArrayList.class, seedDB.scheme.verticalPartitions());
|
|
for (int verticalPosition = 0; verticalPosition < seedDB.scheme.verticalPartitions(); verticalPosition++) {
|
|
seedlists[verticalPosition] = selectVerticalDHTPositions(seedDB, wordhash, minage, Integer.MIN_VALUE, redundancy, verticalPosition);
|
|
}
|
|
return seedlists;
|
|
}
|
|
|
|
/**
|
|
* collecting vertical positions: that chooses for each of the DHT partition a collection of redundant storage positions
|
|
* @param seedDB the database of seeds
|
|
* @param wordhash the word we are searching for
|
|
* @param minage the minimum age of a seed in days (to prevent that too young seeds which cannot have results yet are asked)
|
|
* @param minWordCount the minimum RWI words count of each seed
|
|
* @param redundancy the number of redundant peer position for this parition, minimum is 1
|
|
* @param verticalPosition the verical position, thats the number of the partition 0 <= verticalPosition < seedDB.scheme.verticalPartitions()
|
|
* @return a list of seeds for the redundant positions
|
|
*/
|
|
private static ArrayList<Seed> selectVerticalDHTPositions(final SeedDB seedDB, final byte[] wordhash, final int minage, final int minWordCount, final int redundancy, int verticalPosition) {
|
|
// this method is called from the search target computation
|
|
ArrayList<Seed> seeds = new ArrayList<>(redundancy);
|
|
final long dhtVerticalTarget = seedDB.scheme.verticalDHTPosition(wordhash, verticalPosition);
|
|
final byte[] verticalhash = Distribution.positionToHash(dhtVerticalTarget);
|
|
final Iterator<Seed> dhtEnum = getAcceptRemoteIndexSeeds(seedDB, verticalhash, redundancy, false);
|
|
int c = Math.min(seedDB.sizeConnected(), redundancy);
|
|
int cc = 20; // in case that the network grows rapidly, we may jump to several additional peers but that must have a limit
|
|
while (dhtEnum.hasNext() && c > 0 && cc-- > 0) {
|
|
Seed seed = dhtEnum.next();
|
|
if (seed == null || seed.hash == null) continue;
|
|
if (!seed.getFlagAcceptRemoteIndex()) continue; // probably a robinson peer
|
|
if (seed.getAge() < minage) continue; // prevent bad results because of too strong network growth
|
|
if(seed.getWordCount() < minWordCount) {
|
|
/* Even if the peer is not a robinson and has the required minimum age, it may have an empty or disabled RWI */
|
|
continue;
|
|
}
|
|
if (RemoteSearch.log.isInfo()) RemoteSearch.log.info("selectPeers/DHTorder: " + seed.hash + ":" + seed.getName() + "/ score " + c);
|
|
seeds.add(seed);
|
|
c--;
|
|
}
|
|
return seeds;
|
|
}
|
|
|
|
public static byte[] selectRandomTransferStart() {
|
|
return ASCII.getBytes(Base64Order.enhancedCoder.encode(Digest.encodeMD5Raw(Long.toString(System.currentTimeMillis()))).substring(2, 2 + Word.commonHashLength));
|
|
}
|
|
|
|
public static byte[] limitOver(final SeedDB seedDB, final byte[] startHash) {
|
|
final Iterator<Seed> seeds = getAcceptRemoteIndexSeeds(seedDB, startHash, 1, false);
|
|
if (seeds.hasNext()) return ASCII.getBytes(seeds.next().hash);
|
|
return null;
|
|
}
|
|
|
|
public static List<Seed> getAcceptRemoteIndexSeedsList(
|
|
SeedDB seedDB,
|
|
final byte[] starthash,
|
|
int max,
|
|
boolean alsoMyOwn) {
|
|
final Iterator<Seed> seedIter = getAcceptRemoteIndexSeeds(seedDB, starthash, max, alsoMyOwn);
|
|
final ArrayList<Seed> targets = new ArrayList<Seed>();
|
|
while (seedIter.hasNext() && max-- > 0) targets.add(seedIter.next());
|
|
return targets;
|
|
}
|
|
|
|
/**
|
|
* returns an enumeration of yacySeed-Objects that have the AcceptRemoteIndex-Flag set
|
|
* the seeds are enumerated in the right order according to the DHT
|
|
* @param seedDB
|
|
* @param starthash
|
|
* @param max
|
|
* @param alsoMyOwn
|
|
* @return
|
|
*/
|
|
public static Iterator<Seed> getAcceptRemoteIndexSeeds(final SeedDB seedDB, final byte[] starthash, final int max, final boolean alsoMyOwn) {
|
|
return new acceptRemoteIndexSeedEnum(seedDB, starthash, Math.min(max, seedDB.sizeConnected()), alsoMyOwn);
|
|
}
|
|
|
|
private static class acceptRemoteIndexSeedEnum extends LookAheadIterator<Seed> implements Iterator<Seed>, Iterable<Seed> {
|
|
|
|
private final Iterator<Seed> se;
|
|
private final SeedDB seedDB;
|
|
private int remaining;
|
|
private final boolean alsoMyOwn;
|
|
|
|
private acceptRemoteIndexSeedEnum(SeedDB seedDB, final byte[] starthash, int max, boolean alsoMyOwn) {
|
|
this.seedDB = seedDB;
|
|
this.se = new seedDHTEnum(seedDB, starthash, alsoMyOwn);
|
|
this.remaining = max;
|
|
this.alsoMyOwn = alsoMyOwn;
|
|
}
|
|
|
|
@Override
|
|
protected Seed next0() {
|
|
if (this.remaining <= 0) return null;
|
|
Seed s = null;
|
|
try {
|
|
while (this.se.hasNext()) {
|
|
s = this.se.next();
|
|
if (s == null) return null;
|
|
if (s.getFlagAcceptRemoteIndex() ||
|
|
(this.alsoMyOwn && s.hash.equals(this.seedDB.mySeed().hash)) // Accept own peer regardless of FlagAcceptRemoteIndex
|
|
) {
|
|
this.remaining--;
|
|
break;
|
|
}
|
|
}
|
|
return s;
|
|
} catch (final kelondroException e) {
|
|
System.out.println("DEBUG acceptRemoteIndexSeedEnum:" + e.getMessage());
|
|
Network.log.severe("database inconsistency (" + e.getMessage() + "), re-set of db.");
|
|
this.seedDB.resetActiveTable();
|
|
return null;
|
|
}
|
|
}
|
|
|
|
}
|
|
|
|
private static class seedDHTEnum implements Iterator<Seed> {
|
|
|
|
private Iterator<Seed> e;
|
|
private int steps;
|
|
private final SeedDB seedDB;
|
|
private boolean alsoMyOwn;
|
|
private int pass, insertOwnInPass;
|
|
private Seed nextSeed;
|
|
|
|
private seedDHTEnum(final SeedDB seedDB, final byte[] firstHash, final boolean alsoMyOwn) {
|
|
this.seedDB = seedDB;
|
|
this.steps = seedDB.sizeConnected() + ((alsoMyOwn) ? 1 : 0);
|
|
this.e = seedDB.seedsConnected(true, false, firstHash, yacyVersion.YACY_HANDLES_COLLECTION_INDEX);
|
|
this.pass = 1;
|
|
this.alsoMyOwn = alsoMyOwn;
|
|
if (alsoMyOwn) {
|
|
this.insertOwnInPass = (Base64Order.enhancedCoder.compare(ASCII.getBytes(seedDB.mySeed().hash), firstHash) > 0) ? 1 : 2;
|
|
} else {
|
|
this.insertOwnInPass = 0;
|
|
}
|
|
this.nextSeed = nextInternal();
|
|
}
|
|
|
|
@Override
|
|
public boolean hasNext() {
|
|
return (this.nextSeed != null) || this.alsoMyOwn;
|
|
}
|
|
|
|
public Seed nextInternal() {
|
|
if (this.steps == 0) return null;
|
|
this.steps--;
|
|
|
|
if (!this.e.hasNext() && this.pass == 1) {
|
|
// rotate from the beginning; this closes the ordering of the DHT at the ends
|
|
this.e = this.seedDB.seedsConnected(true, false, null, yacyVersion.YACY_HANDLES_COLLECTION_INDEX);
|
|
this.pass = 2;
|
|
}
|
|
if (this.e.hasNext()) {
|
|
return this.e.next();
|
|
}
|
|
this.steps = 0;
|
|
return null;
|
|
}
|
|
|
|
@Override
|
|
public Seed next() {
|
|
if (this.alsoMyOwn &&
|
|
((this.pass > this.insertOwnInPass) ||
|
|
(this.pass == this.insertOwnInPass && this.nextSeed == null) || // Own hash is last in line
|
|
(this.pass == this.insertOwnInPass && this.nextSeed != null && (Base64Order.enhancedCoder.compare(ASCII.getBytes(this.seedDB.mySeed().hash), ASCII.getBytes(this.nextSeed.hash)) < 0)))
|
|
) {
|
|
// take my own seed hash instead the enumeration result
|
|
this.alsoMyOwn = false;
|
|
return this.seedDB.mySeed();
|
|
}
|
|
final Seed next = this.nextSeed;
|
|
this.nextSeed = nextInternal();
|
|
return next;
|
|
}
|
|
|
|
@Override
|
|
public void remove() {
|
|
throw new UnsupportedOperationException();
|
|
}
|
|
}
|
|
|
|
/**
|
|
* enumerate peers that provide remote crawl urls
|
|
* @param seedDB
|
|
* @return an iterator of seed objects
|
|
*/
|
|
public static Iterator<Seed> getProvidesRemoteCrawlURLs(final SeedDB seedDB) {
|
|
return new providesRemoteCrawlURLsEnum(seedDB);
|
|
}
|
|
|
|
private static class providesRemoteCrawlURLsEnum extends LookAheadIterator<Seed> implements Iterator<Seed>, Iterable<Seed> {
|
|
|
|
private final Iterator<Seed> se;
|
|
private final SeedDB seedDB;
|
|
|
|
private providesRemoteCrawlURLsEnum(final SeedDB seedDB) {
|
|
this.seedDB = seedDB;
|
|
this.se = seedDB.seedsConnected(true, false, null, yacyVersion.YACY_POVIDES_REMOTECRAWL_LISTS);
|
|
}
|
|
|
|
@Override
|
|
protected Seed next0() {
|
|
Seed s;
|
|
try {
|
|
while (this.se.hasNext()) {
|
|
s = this.se.next();
|
|
if (s == null) return null;
|
|
if (s.getLong(Seed.RCOUNT, 0) > 0) return s;
|
|
}
|
|
} catch (final kelondroException e) {
|
|
System.out.println("DEBUG providesRemoteCrawlURLsEnum:" + e.getMessage());
|
|
Network.log.severe("database inconsistency (" + e.getMessage() + "), re-set of db.");
|
|
this.seedDB.resetActiveTable();
|
|
return null;
|
|
}
|
|
return null;
|
|
}
|
|
|
|
}
|
|
|
|
/**
|
|
* get either the youngest or oldest peers from the seed db. Count as many as requested
|
|
* @param seedDB
|
|
* @param up if up = true then get the most recent peers, if up = false then get oldest
|
|
* @param count number of wanted peers
|
|
* @param minWordCount the minimum RWI words count of each seed
|
|
* @return a hash map of peer hashes to seed object
|
|
*/
|
|
public static ConcurrentMap<String, Seed> seedsByAge(final SeedDB seedDB, final boolean up, int count, final int minWordCount) {
|
|
if (count > seedDB.sizeConnected()) count = seedDB.sizeConnected();
|
|
Seed ys;
|
|
final Iterator<Seed> seeds = seedDB.seedsSortedConnected(!up, Seed.LASTSEEN);
|
|
try {
|
|
final ConcurrentMap<String, Seed> result = new ConcurrentHashMap<String, Seed>();
|
|
while (seeds.hasNext() && count-- > 0) {
|
|
ys = seeds.next();
|
|
if (ys != null && ys.hash != null && ys.getWordCount() >= minWordCount) {
|
|
result.put(ys.hash, ys);
|
|
}
|
|
}
|
|
return result;
|
|
} catch (final kelondroException e) {
|
|
Network.log.severe("Internal Error at yacySeedDB.seedsByAge: " + e.getMessage(), e);
|
|
return null;
|
|
}
|
|
}
|
|
|
|
/**
|
|
* get either the youngest or oldest peers from the seed db. Count as many as requested
|
|
* @param seedDB
|
|
* @param up if up = true then get the most recent peers, if up = false then get oldest
|
|
* @param count number of wanted peers
|
|
* @return a hash map of peer hashes to seed object
|
|
*/
|
|
public static ConcurrentMap<String, Seed> seedsByAge(final SeedDB seedDB, final boolean up, int count) {
|
|
return seedsByAge(seedDB, up, count, Integer.MIN_VALUE);
|
|
}
|
|
|
|
} |