The storage proxy needs to collect statistics about read, write and
range. For that the ihistogram object was added to its stats object.
Signed-off-by: Amnon Heiman <amnon@cloudius-systems.com>
Make storage proxy a singleton and add helpers to look up a reference.
This makes it easier to convert code from Origin in areas such as
storage service.
Signed-off-by: Pekka Enberg <penberg@cloudius-systems.com>
"This series adds the storage proxy metrics API. The definition are based on the
StorageProxyMetrics definition. This series also adds stats object to the
storage_proxy a getter function for it and an implementation based on it, but
it is currently does not adds the code to manipulate the counters."
This adds a stat object with counters that will be used by the API. The
stat object instance will be return with a get_stats method.
Signed-off-by: Amnon Heiman <amnon@cloudius-systems.com>
In Origin, partitions in range query results are ordered using
decorated_key ordering. This allows the use of token() function for
incremental iterating over results, as mentioned here:
http://www.datastax.com/dev/blog/client-side-improvements-in-cassandra-2-0
We may also need this to implement paging.
The old code didn't preserve ordering, because it didn't merge-sort
data coming from different shards. The fix relies on
query_mutations_locally(), which already preserves the ordering. We're
going to use mutation queries for range queries anyway.
Works only if all replicas (participating in CL) has the same live
data. Does not detects mismatch in tombstones (no infrastructure yet).
Does not report timeout yet.
If local mutation write takes longer then write timeout mutation will
be deleted while it is processed by database engine. Fix this by storing
mutation in shared pointer and hold to the pointer until mutation is
locally processed.
This reverts commit 52aa0a3f91.
After c9909dd183 this is no longer needed since reference to a
handler is not used in abstract_write_response_handler::wait() continuation.
Conflicts:
service/storage_proxy.cc
If last response comes after write timeout is triggered, but before
continuation, that suppose to handle it runs the handler can be removed
to earlier and be access from the continuation after deletion. Fix it by
making response handler to be shared pointer instead of unique and
holding to it in timeout continuation.
partitions_ranges will be manipulated upon to be split for different
destination, so provide it separately from read_command to not copy the
later for each destination.
Don't return foreign_ptr<> which is not copyable so that we can use
map_difference for maps with result_set in them.
Signed-off-by: Pekka Enberg <penberg@cloudius-systems.com>
The immediate motivation for introducing frozen_mutation is inability
to deserialize current "mutation" object, which needs schema reference
at the time it's constructed. It needs schema to initialize its
internal maps with proper key comparators, which depend on schema.
frozen_mutation is an immutable, compact form of a mutation. It
doesn't use complex in-memory strucutres, data is stored in a linear
buffer. In case of frozen_mutation schema needs to be supplied only at
the time mutation partition is visited. Therefore it can be trivially
deserialized without schema.
Partitions should be ordered using Origin's ordering, which is first
by token, then by Origin's representation of the key. That is the
natural ordering of decorated_key.
This also changes mutation class to hold decorated_key, to avoid
decoration overhead at different layers.
This patch converts (for very small value of 'converts') some
replication related classes. Only static topology is supported (it is
created in keyspace::create_replication_strategy()). During mutation
no replication is done, since messaging service is not ready yet,
only endpoints are calculated.
s/database/distributed<database>/ everywhere.
Use simple distribution rules: writes are broadcast, reads are local.
This causes tremendous data duplication, but will change soon.
