This changes the API of the types but does not alter the functionality
since only the `context.Background()` is currently being used. Adding
`Context` will help handle fault tolerance in the future by allowing
timeouts to be propagated across calls to different nodes.
A `StatsClient` for `expvar` is added so we can track Stats through
the `/debug/vars` endpoint. Tags are nested inside maps so that we
can see stats for db, frame & slice.
Also added a `MultiStatsClient` for chaining multiple `StatsClient`
implementations together (e.g. `expvar` and DataDog).
Introduces new `pilosactl sort` to sort import files by bit
position so they can be inserted faster. Also optimizes container
scanning and adds a `-buffer-size` flag to `import`.
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.
CSV exports can now be done with the pilosactl application:
$ pilosactl -d mydb -f myframe -o MYFILE.csv
If `-o` is not specified then the CSV is written to STDOUT. The
exporter combines all slices for the db/frame to into a single
concatenated CSV file.
This commit adds `roaring.Bitmap.CountRange()` to return the number
of bits in a subrange of a bitmap. This significantly improves
server start time and is needed for upcoming zero copy bitmap
optimizations.
This commit moves the cache flush to the `Index` and only serializes
a single fragment at a time.
Also included in this commit is the `inspect` command for the
`pilsoactl` binary. This provides insight into pilosa data files.
This commit refactors the block checksumming by removing the
iteration over each block and instead only checking blocks which
have data. This requires merging the cache inspection with the
roaring iterator to reduce CPU time and memory allocations.
This commit adds a `Closing` channel to the `IndexSyncer` and the
`FragmentSyncer` that is periodically checked during execution.
If the channel is closed then an in-progress sync is immediately
stopped.
This commit changes the fragment allocation algorithm in the cluster
to make use of the `DB` name. This allows each database to use a
different slice distribution.
Initially, the `frame` was going to be used for allocation, however,
this was problematic since queries can span multiple frames so it's
impossible to choose a single frame to use.
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 refactors the anti-entropy system to fetch data from
all replicated blocks and only set/clear bits which deviate from
the consensus between all blocks.
An example of this is if 3 nodes had the following bits set for
a single bitmap:
Node A: 1 2 3
Node B: 2 4
Node C: 1 2 4
Then only bits which are set on a majority will be set. In this
case bits 1, 2, & 4 are set but 3 only exists on a single node.
The node performing the merge would then determine the following
set/clear diffs for each node:
Node A: clear(3), set(4)
Node B: set(1)
Node C: none
Once the merge is performed and all nodes receive their diff
instructions then the nodes will be in sync:
Node A: 1 2 4
Node B: 1 2 4
Node C: 1 2 4
There still exists situations where bits can be reset. If Node A
is up and Node B & C are down then Node A's bits will be reset
once B & C come back online. We should add write consistency
settings for incoming writes so that we can ensure that a quorum
is written to before returning a success. This is outside the
scope of this commit though.
