NAME
Data::CountingBloomFilter::Shared - shared-memory counting Bloom filter for Linux
SYNOPSIS
use Data::CountingBloomFilter::Shared;
# sized for 1_000_000 items at a 1% false-positive rate, anonymous mapping
my $cbf = Data::CountingBloomFilter::Shared->new(undef, 1_000_000, 0.01);
$cbf->add("alice");
$cbf->add("bob");
$cbf->contains("alice"); # 1 (probably present)
$cbf->contains("carol"); # 0 (definitely absent)
# unlike a plain Bloom filter, you can delete
$cbf->remove("alice");
$cbf->contains("alice"); # 0
# ...and read an occurrence count (0..15): how many times an item is stored
$cbf->add("x"); $cbf->add("x");
$cbf->count_of("x"); # 2
# bulk add in a single lock acquisition
my $n = $cbf->add_many([ map { "user-$_" } 1 .. 1000 ]);
# share across processes via a backing file
my $shared = Data::CountingBloomFilter::Shared->new("/tmp/seen.cbf", 1_000_000);
# freeze and ship: query it read-only (lock-free) on other machines
$shared->freeze;
my $ro = Data::CountingBloomFilter::Shared->new_readonly("/tmp/seen.cbf");
$ro->contains("alice");DESCRIPTION
A counting Bloom filter in shared memory: like Data::BloomFilter::Shared, a compact fixed-size structure for approximate set membership, but each position is a small 4-bit counter instead of a single bit. That one change buys two things a plain Bloom filter cannot do: you can remove items, and you can ask how many times an item was added (count_of). The cost is memory -- four bits per slot instead of one, so about four times the size of the equivalent Bloom filter.
Membership is still one-sided: contains returns "definitely not present" or "probably present". For items you have added (and not removed) it always returns true -- there are no false negatives -- with a small tunable rate of false positives. It never stores the items themselves, only which counters they touch.
Each item is hashed once with XXH3 (128-bit) and, by double hashing (Kirsch-Mitzenmacher), drives k probes into an array of m counters. add increments each of the item's k counters (saturating at 15); contains is true when all k are greater than zero; remove decrements them (only if the item is present); and count_of returns the minimum of the k counters -- an estimate of the item's stored occurrence count. From the requested capacity n and false-positive rate p the filter derives k = round(-log2 p) and m = next_pow2(n * k / ln2), the same geometry as a Bloom filter.
The counters saturate at 15: an item added more than 15 times (or colliding with others up to that ceiling) sticks at 15 and is never decremented again, which keeps membership sound (no false negatives) but caps count_of and means a saturated item cannot be fully removed. Sizing the filter for its intended load keeps saturation vanishingly rare.
Because the table lives in a shared mapping, several processes share one filter: any process that opens the same backing file, inherits the anonymous mapping across fork, or reopens a passed memfd sees the others' additions and removals and contributes its own. A write-preferring futex rwlock with dead-process recovery guards mutation, so many processes may add, remove, and contains concurrently.
Removal caveat. remove decrements the counters of an item that is present. Because counters are shared between items, decrementing the counters of an item that was never added -- or one whose probes collide with present items -- can push a shared counter to zero and cause a false negative for some other item. Only remove items you actually added, and remove an item as many times as you added it to forget it completely.
Items are added, tested, and removed by their byte content; wide-character strings (any codepoint above 255) cause a "Wide character" croak -- encode such strings to bytes first (for example with Encode::encode_utf8). Linux-only. Requires 64-bit Perl.
METHODS
Constructors
my $cbf = Data::CountingBloomFilter::Shared->new($path, $capacity, $fp_rate);
my $cbf = Data::CountingBloomFilter::Shared->new(undef, 1_000_000); # anonymous, 1% default
my $cbf = Data::CountingBloomFilter::Shared->new_memfd($name, $capacity, $fp_rate);
my $cbf = Data::CountingBloomFilter::Shared->new_from_fd($fd);
my $ro = Data::CountingBloomFilter::Shared->new_readonly($path); # frozen file, read-only$path is the backing file (undef or omitted for an anonymous mapping). $capacity is the number of items you expect to add (at least 1). $fp_rate is the target false-positive rate at that capacity, strictly between 0 and 1 (default 0.01). new and new_memfd croak on a capacity below 1 or an out-of-range $fp_rate.
From $capacity and $fp_rate the filter derives k = round(-log2 fp_rate) (clamped to 1..32) probes and m = next_pow2(capacity * k / ln2) 4-bit counters (floor 64), for a m/2-byte counter array. When reopening an existing file or memfd the stored geometry wins, so the caller's $capacity/$fp_rate do not resize it -- but they are still range-checked first, so an out-of-range value croaks even though it would have been ignored. new_memfd creates a Linux memfd (transferable via its memfd descriptor); new_from_fd reopens one in another process. The descriptor you pass is duplicated (F_DUPFD_CLOEXEC), so it stays yours to close and closing it does not disturb the handle.
An optional file mode may be passed as the last argument to new (e.g. 0660) to opt a newly-created backing file into cross-user sharing; it defaults to 0600 (owner-only) and is ignored for anonymous mappings and existing files. new_readonly opens a frozen file read-only for lock-free querying (see "FROZEN (READ-ONLY) MODE").
Adding, testing, counting, removing
my $new = $cbf->add($item); # 1 if the item was probably new, else 0
my $added = $cbf->add_many(\@items); # count of adds that were probably new
my $in = $cbf->contains($item); # 1 if probably present, 0 if definitely absent
my $c = $cbf->count_of($item); # occurrence estimate 0..15
my $gone = $cbf->remove($item); # 1 if present and decremented, else 0
$cbf->clear; # reset to emptyadd hashes $item (by its bytes; wide characters croak, encode first) and increments its k counters, each saturating at 15. It returns 1 if the item was probably new (at least one of its counters was 0 beforehand) or 0 if it was already present. add_many takes an array reference and does the whole batch under a single write lock, returning how many of the adds were probably new.
contains returns 1 if the item is probably present (all k counters are nonzero) and 0 if it is definitely absent. A 0 means definitely absent: an item you added and have not removed never returns 0 (no false negatives). A 1 may be a false positive.
count_of returns the minimum of the item's k counters, an integer from 0 to 15 estimating how many times the item is stored (times added minus removed). Collisions can only raise a counter, so below saturation count_of never under-counts -- it is an upper estimate. It saturates at 15: a returned 15 means 15 or more, so an item added more than 15 times is under-reported. A 0 means definitely absent.
remove, only if the item is present (all k counters are nonzero), decrements each of them and returns 1; otherwise it changes nothing and returns 0. Saturated (15) counters are left stuck, so remove of a saturated item still returns 1 but cannot lower it -- a saturated item cannot be fully removed. See the Removal caveat in "DESCRIPTION": only remove items you added, and remove an item as many times as it was added. clear empties the whole filter (all counters zeroed).
Merging
$cbf->merge($other); # counter-wise saturating addmerge adds another filter's counters into this one, counter by counter, saturating each at 15. Both filters must have the same geometry (same m and k, i.e. created with the same capacity and false-positive rate) or merge croaks. The other filter is snapshotted under its own read lock, so two processes may safely merge concurrently. After merging, contains is true for every item present in either filter and count_of reflects the summed (saturated) counts.
Introspection and lifecycle
$cbf->count; $cbf->capacity; $cbf->counters; $cbf->hashes; $cbf->fp_rate;
$cbf->stats; $cbf->path; $cbf->memfd; $cbf->sync; $cbf->unlink;count estimates the number of distinct items currently present (added minus removed, from the fraction of nonzero counters, -(m/k) * ln(1 - X/m) where X is the nonzero-counter count); it is an estimate, not an exact tally. capacity is the configured item capacity; counters is the counter count m (a power of two); hashes is k; fp_rate is the configured target false-positive rate. sync flushes the mapping to its backing store (a no-op for anonymous and memfd filters); unlink removes the backing file (also callable as Class->unlink($path)) and croaks if the removal fails -- except when the file is already gone, which is what you asked for; it is likewise a no-op when there is no backing file (anonymous or memfd); path returns the backing path (undef for anonymous, memfd, or fd-reopened filters) and memfd the backing descriptor -- the memfd of a new_memfd filter or the dup'd fd of a new_from_fd filter, and -1 for file-backed or anonymous filters.
STATS
stats() returns a hashref describing the filter:
capacity-- the configured item capacity.fp_rate-- the configured target false-positive rate.counters-- the counter countm(a power of two).hashes-- the number of probeskper item.counters_set-- the number of nonzero counters.count-- the estimated number of distinct items added.fill_ratio--counters_set / counters, between 0 and 1. As this approaches 1 the filter is saturating and the false-positive rate degrades.ops-- running count of write-path calls (add,add_many,remove,merge,clear), whether or not any counter actually changed.mmap_size-- bytes of the shared mapping.frozen-- 1 if the filter has been sealed byfreeze(immutable), else 0.readonly-- 1 if this handle is a read-only view (fromnew_readonly, or the handle that calledfreeze), else 0.
SHARING ACROSS PROCESSES
The filter lives in a shared mapping, shared the same three ways as the rest of the family: a backing file (every process calls new($path, ...) on the same path with a matching capacity), an anonymous mapping inherited across fork, or a memfd whose descriptor is passed to an unrelated process (over a UNIX socket via SCM_RIGHTS, or via /proc/$pid/fd/$n) and reopened with new_from_fd($fd). Because the mapping is shared, every process adds into, tests against, and removes from the same table.
# producer and consumer share one filter with no coordination
my $cbf = Data::CountingBloomFilter::Shared->new(undef, 100_000); # before fork
unless (fork) { $cbf->add_many([ map { "ev-$_" } 1 .. 1000 ]); exit }
wait;
print $cbf->contains("ev-500") ? "seen\n" : "no\n"; # seen -- the child's addFROZEN (READ-ONLY) MODE
A file-backed filter can be frozen and then shipped to other machines, where consumers open it read-only and query it with no locking at all.
# producer: build, freeze, ship the file
my $cbf = Data::CountingBloomFilter::Shared->new("/tmp/seen.cbf", 1_000_000, 0.01);
$cbf->add_many(\@known);
$cbf->freeze; # seal: now immutable, and $cbf itself is read-only
# ... copy /tmp/seen.cbf to another host ...
# consumer (any process, same architecture): read-only, lock-free
my $ro = Data::CountingBloomFilter::Shared->new_readonly("/tmp/seen.cbf");
$ro->contains($item) for @queries;
$ro->count_of($item);freeze takes the write lock, marks the filter permanently immutable (there is no unfreeze -- rebuild the file to change it), and flushes the seal to disk. A frozen filter rejects every mutator (add, add_many, remove, merge, clear) with a croak, and a read-write reopen (new($path, ...)) of a sealed file is refused -- so a shipped artifact can never be silently mutated out from under its readers. That protection is enforced by the reader: the seal is a header flag that 0.01 and earlier do not know about, and the on-disk format version is deliberately unchanged so those releases can still open files written here. A pre-0.02 build therefore opens a sealed file read-write and can modify it, so keep producers and consumers on 0.02 or later if you rely on the seal. freeze itself is not idempotent: the handle that seals the file becomes a read-only view of it, so calling freeze on that handle again croaks.
new_readonly($path) maps the file O_RDONLY / PROT_READ and requires it to be frozen (it croaks on a file that was never freezed). Because a sealed filter's counters and geometry are immutable, contains, count_of, count, and stats read them directly, taking no reader lock -- the mapping is never written, so a read-only view works from a read-only file descriptor or a read-only filesystem, and any number of processes can share one PROT_READ mapping. frozen and readonly report the two states.
Portability. The on-disk format is native binary (native-endian 64-bit words), so a frozen file may be copied only between machines of the same architecture; a wrong-endian file is rejected at open by the magic check. Copy the file to each consumer -- do not share one file over a network filesystem: the lock is a Linux futex (process-local to one kernel), and the "no live writer" contract assumes a static copy. Linux-only; 64-bit Perl.
SECURITY
Backing files are created with mode 0600 (owner-only) by default, so only the creating user can open and attach them. To share a backing file across users, pass an explicit octal file mode such as 0660 as the last argument to new; the mode is applied when the file is created; a pre-existing file owned by the caller is adopted as new -- and likewise gets the requested mode -- when it is empty, or when it is the full-size all-zero file an interrupted create leaves behind (see "CRASH SAFETY"). Any other existing file keeps its own permissions. The file is opened with O_NOFOLLOW, so a symlink planted at the path is refused, and created with O_EXCL; the on-disk header is validated when the file is attached. Any process you grant write access to a shared mapping is trusted not to corrupt its contents while other processes are using it.
CRASH SAFETY
Mutation is guarded by a futex-based write-preferring rwlock with PID-encoded ownership; if a holder dies, the next contender detects the dead owner and recovers. Each add and remove is a short sequence of counter updates, so a crash leaves the filter consistent up to the last completed operation. Limitation: PID reuse is not detected (very unlikely in practice).
Reader-slot exhaustion (slotless readers): dead-process recovery attributes a crashed lock holder's contribution through its reader-slot. The slot table holds 1024 entries (one per concurrent reader process). If more than that many reader processes share one mapping at once, a reader that cannot claim a slot proceeds "slotless" -- it still takes the read lock but leaves no per-process record. If such a slotless reader is then killed while holding the read lock, its share of the lock cannot be attributed to a dead process, so writer recovery cannot reclaim it and writers may block until the mapping is recreated. Reaching this needs more than 1024 concurrent reader processes on one mapping plus a crash in the brief read-lock window; the dead-process slot reclaim keeps the table from filling with stale entries, so in practice it is very unlikely. Those preconditions cover the live-process route only. The count lives in the mapping and new validates the geometry, not this transient value, so a backing file damaged at rest -- bit rot, a partial copy, or a process that scribbled on the mapping -- can present a non-zero slotless count and block every writer the same way, with none of the above. If writers hang on a file no live reader is using, recreate it.
An interrupted create is recovered too. A creator killed after the backing file is sized but before its header is committed leaves a full-size, all-zero file. new re-initializes such a file automatically, but only when it is exactly the size the requested geometry needs, is owned by your effective uid, and is still entirely zero -- a file holding data is never re-initialized. If the creator got as far as writing part of the header, the file cannot be told apart from a corrupt one and new croaks with incomplete counting Bloom filter file left by an interrupted create; remove it and retry. A file left behind by an interrupted create never held data, so removing it is safe -- but a file whose header was corrupted after the fact reaches the same croak, so confirm it is an abandoned create before deleting anything you care about.
Disk space. The backing file is created sparse: new sizes it, but blocks are allocated only as you write, so a large filter costs almost nothing on disk until it is used. The cost of that is a late failure, and how it reaches you depends on the filesystem. Where blocks are allocated at fault time -- tmpfs, so /dev/shm and many /tmp mounts -- a write to a page that cannot be backed raises SIGBUS and kills the process, because an mmap store has no way to report ENOSPC. Where allocation is delayed to writeback (ext4, xfs), the store lands in page cache and the failure appears later: the write is lost, and sync is what reports it, croaking with the underlying error. Keep the filesystem sized for the filter you asked for, and call sync when you need to know your writes reached disk.
SEE ALSO
Data::BloomFilter::Shared (membership without delete, one bit per slot), Data::CuckooFilter::Shared (membership with delete and count_of, no saturation caveat), Data::HyperLogLog::Shared, and the rest of the Data::*::Shared family.
AUTHOR
vividsnow
LICENSE
This is free software; you can redistribute it and/or modify it under the same terms as Perl itself.