Previously the index would create a fragment if it didn't exist
when clearing a bit. This causes additional file handles to be
created when it's already known that the bit is not set.
Previously the `slice` was not copied to each new segment when
`Intersect()`, `Union()`, and `Difference()` operations were performed.
This causes issues when those bitmaps were then operated on later such
as performing a `TopN()` on a source bitmap.
This commit makes several changes to optimize the TopN() query:
- Reduce highbits() back from 24-bits to 16-bits.
- Reduce MaxArraySize back from 2^20 to 4096.
- Optimize bitmap count invalidation.
- Parallelize TopN() across nodes.
- Parallelize TopN() across slices.
This commit moves bitmap intersection into the `roaring` package
in order to reduce allocations.
Benchmarks against real world datasets show a speed improvement of
353%. Previously, intersection operated at approximately 5.6M
bits/sec against two 60K bit bitmaps with 30% overlap. The new
intersection operates at approximately 20M bits/sec.
This commit moves the computation of the intersection count to
the `roaring` package so that no allocations are required. The
implementation operates at the roaring container level and has
specialized functions for array-array, array-bitmap, and
bitmap-bitmap container pairs.
This commit changes the underlying storage for `Bitmap` from a
red-black tree to a roaring bitmap. It also removes bitmaps from
the cache and only stores the bitmap count.
This commit adds the ability to set string, integer, and boolean
values on bitmaps and profiles within Pilosa. Bitmap attributes
are automatically returned when making a `Bitmap()` call. Profile
attributes must be requested by setting `profile=true` in the
URL.
The function names have also been renamed to initial caps so that
the PQL query language can support math operations in the future.
This commit adds integration testing for the `cmd/pilosa` package.
It uses randomly generated `set()` data and verifies the data using `get()`
both immediately after the insert and after restart.
This commit attaches the mmap-backed roaring bitmaps as storage
to the fragments. Currently snapshotting is not supported from
Pilosa so the WAL will grow indefinitely, however, it is not a
difficult change to make.
This commit refactors the pilosa codebase. It makes several major
changes:
* Removes bitmap handles
* Removes dispatch/hold/transport
* Removes etcd dependency
* Adds consistent hash ring for slice placement
* Refactors parser/lexer
* Adds strong typing to PQL AST
* Flattens package hierarchy
This commit moves the index package to the root. Because pilosa
is an index at its core, it's redundant to have an index
subpackage. It also provides better naming such as `pilosa.Bitmap`
instead of `index.Bitmap`.
