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.
This commit adds a simple benchmarking utility to the `pilosactl`
binary. It currently only supports individual `SetBit()` commands
but it's a good start towards making a generic benchmarking
framework at the integration level.
The subcommands and usage/help messages were also cleaned up to
output correctly.
This commit changes the backup format of a fragment from a
simple file stream to a tar archive which combines the data file
and the cache file.
Fixes#63
This commit refactors the profile and bitmap attribute stores so
that they share the same code. There is a new `AttrStore` which
associates key/value pairs with a `uint64` identifier.
The previous JSON encoding has been fixed to use protobufs which
fixes encoding issues for int64.
This commit adds the ability of the Fragment to flush the cache
bitmap IDs to disk periodically. They can then be reloaded when
the fragment is reopened.
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 refactors the roaring bitmaps to use uint64 values
instead of uint32 values. This is required for the size of
values we need in pilosa.
This change is not backwards compatible with the previous data
format so any existing data directories need to be removed before
using this code.
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 simplifies Fragment by replacing the request/response channels
with a `sync.Mutex` to restrict access. This also simplifies the
`FragmentContainer` code which can largely be removed now since most of
the methods are simply wrappers around `GetFragment()` and `Fragment`.
