Originally, streamed_mutations guaranteed that emitted tombstones are
disjoint. In order to achieve that two separate objects were produced
for each range tombstone: range_tombstone_begin and range_tombstone_end.
Unfortunately, this forced sstable writer to accumulate all clustering
rows between range_tombstone_begin and range_tombstone_end.
However, since there is no need to write disjoint tombstones to sstables
(see #1153 "Write range tombstones to sstables like Cassandra does") it
is also not necessary for streamed_mutations to produce disjoint range
tombstones.
This patch changes that by making streamed_mutation produce
range_tombstone objects directly.
Signed-off-by: Paweł Dziepak <pdziepak@scylladb.com>
Checking bloom filters of sstables to compute max purgeable timestamp
for compaction is expensive in terms of CPU time. We can avoid
calculating it if we're not about to GC any tombstone.
This patch changes compacting functions to accept a function instead
of ready value for max_purgeable.
I verified that bloom filter operations no longer appear on flame
graphs during compaction-heavy workload (without tombstones).
Refs #1322.
In order to support ByteOrderedPartitioner, we need to implement the
missing describe_ownership and midpoint function in
byte_ordered_partitioner class.
As a starter, this path uses a simple node token distance based method
to calculate ownership. C* uses a complicated key samples based method.
We can switch to what C* does later.
Tests are added to tests/partitioner_test.cc.
Fixes#1378
sstable_list is now a map<generation, sstable>; change it to a set
in preparation for replacing it with sstable_set. The change simplifies
a lot of code; the only casualty is the code that computes the highest
generation number.
This patch adds support for thrift prepared statements. It specializes
the result_message::prepared into two types:
result_message::prepared::cql and result_message::prepared::thrift, as
their identifiers have different types.
Signed-off-by: Duarte Nunes <duarte@scylladb.com>
Add contiguity flag to cache entry and set it in scanning reader.
Partitions fetched during scanning are continuous
and we know there's nothing between them.
Clear contiguity flag on cache entries
when the succeeding entry is removed.
Use continuous flag in range queries.
Don't go do disk if we know that there's nothing
between two entries we have in cache. We know that
when continuous flag of the first one is set to true.
Signed-off-by: Piotr Jastrzebski <piotr@scylladb.com>
Message-Id: <72bae432717037e95d1ac9465deaccfa7c7da707.1466627603.git.piotr@scylladb.com>
From Paweł:
This series introduces streaming_mutations which allow mutations to be
streamed between the producers and the consumers as a series of
mutation_fragments. Because of that the mutation streaming interface
works well with partitions larger than available memory provided that
actual producer and consumer implementations can support this as well.
mutation_fragments are the basic objects that are emitted by
streamed_mutations they can represent a static row, a clustering row,
the beginning and the end of a range tombstone. They are ordered by their
clustering keys (with static rows being always the first emitted mutation
fragment). The beginning of range tombstone is emitted before any
clustering row affected by that tombstone and the end of range tombstone
is emitted after the last clustering row affected by it. Range tombstones
are disjoint.
In this series all producers are converted to fully support the new
interface, that includes cache, memtables and sstables. Mutation queries
and data queries are the only consumers converted so far.
To minimize the per-mutation_fragment overhead streamed_mutations use
batching. The actual producer implementation fills a buffer until
it is full (currently, buffer size is 16, the limit should, however,
be changed to depend on the actual size in memory of the stored elements)
or end of stream is reached.
In order to guarantee isolation of writes reads from cache and memtable
use MVCC. When a reader is created it takes a snapshot of the particular
cache or memtable entry. The snapshot is immutable and if there happen
to be any incoming writes while the read is active a new version of
partition is created. When the snapshot is destroyed partition versions
are merged together as much as possible.
Performance results with perf_simple_query (median of results with
duration 15):
before after diff
write 618652.70 618047.58 -0.10%
read 661712.44 608070.49 -8.11%
From Glauber:
This is my new take at the "Move throttler to the LSA" series, except
this one don't actually move anything anywhere: I am leaving all
memtable conversion out, and instead I am sending just the LSA bits +
LSA active reclaim. This should help us see where we are going, and
then we can discuss all memtable changes in a series on its own,
logically separated (and hopefully already integrated with virtual
dirty).
[tgrabiec: trivial merge conflicts in logalloc.cc]
It is incorrect to update row_cache with a memtable that is also its
underlying storage. The reason for that is that after memtable is merged
into row_cache they share lsa region. Then when there is a cache miss
it asks underlying storage for data. This will result with memtable
reader running under row_cache allocation section. Since memtable reader
also uses allocation section the result is an assertion fault since
allocation sections from the same lsa region cannot be nested.
Signed-off-by: Paweł Dziepak <pdziepak@scylladb.com>
With streamed_mutations a partition with many small rows doesn't stress
the cache as much as the test expects. Use large clustering rows instead.
Signed-off-by: Paweł Dziepak <pdziepak@scylladb.com>
"Reclaiming many segments was observed to cause up to multi-ms
latency. With the new setting, the latency of reclamation cycle with
full segments (worst case mode) is below 1ms.
I saw no difference in throughput in a CQL write micro benchmark
in neither of these workloads:
- full segments, reclaim by random eviction
- sparse segments (3% occupancy), reclaim by compaction and no eviction
Fixes #1274."
push_back() is not reentrant with pop_front(), used by the evictor. If
reclaimer runs when std::deque allocates a new node it will get
corrupted. Fix by runnning push_back() under reclaim lock.
"Correctness of current uses of clear() and invalidate() relies on fact
that cache is not populated using readers created before
invalidation. Sstables are first modified and then cache is
invalidated. This is not guaranteed by current implementation
though. As pointed out by Avi, a populating read may race with the
call to clear(). If that read started before clear() and completed
after it, the cache may be populated with data which does not
correspond to the new sstable set.
To provide such guarantee, invalidate() variants were adjusted to
synchronize using _populate_phaser, similarly like row_cache::update()
does.
Fixes #1291."
