GridGain In-Memory Data Fabric Professional Edition Release Notes
=======================================

GridGain In-Memory Data Fabric Professional Edition 2.1.11
------------------------------------
* Fixed array serialization on big-endian architectures.
* Fixed future hang on non-serializable cache response.
* Fixed issue with incorrect partition state recovery leading to "Page is corrupted" message on node restart.
* Fixed MBeans names escaping to handle special symbols.
* Added custom thread pools monitoring.
* Added Spring 3.x integration module.
* Changed Log4J and Log4J2 adapters to update active logger configuration when configuration source file is modified.
* Ensure that SocketChannel is always closed during unsuccessful TCP client connection attempts.
* Opportunistically reopen caches after client node rejoins topology.
* Use PeerDeployAware for stream transformer.
* Visor CMD: Added dialog to put/get/remove values to/from cache for simple key types.
* Web Console: Fixed creation of snapshot operation schedules.
* Web Console: Implemented component for cluster selection and activation.
* Web Console: Improved export of grouped data on Admin panel.

GridGain In-Memory Data Fabric Professional Edition 2.1.10
------------------------------------
* Backup node saves entry on get() from primary if it's not presented locally and local store is used.
* Fixed IgniteOOM propagation to client in case of implicit transaction and one phase commit.
* Fixed NullPointerException during early access to services().
* Fixed resetting non-valid near cache entry.
* Stop ping if current node is stopping.

GridGain In-Memory Data Fabric Professional Edition 2.1.9
------------------------------------
* C++ node can start without example folder now.
* Fixed deadlock when partition eviction starts from multiple threads.
* Fixed bug in cache start operations ordering.
* Fixed PDS corruption if partition have been evicted and owned again.
* Fixed: throttling is not applied to page allocation.
* Fix: when PDS is enabled and IGNITE_BINARY_SORT_OBJECT_FIELDS property is set and IgniteCache#put is called a node hangs.
* Fix for JobStealingCollisionSpi can sends jobs to a node that joined after task was executed.
* Fixed redundant invocations of CacheStoreSessionListener if store operation is not executed.
* S3 IP finder: support server side encryption.
* S3 IP finder: support bucket endpoint configuration.
* Fixed NPE in TcpDiscoverySharedFsIpFinder.
* Ignite Logger type is displayed during Node startup.
* Visor CMD: Added ability to start cache rebalance.
* Visor CMD: Added "-quiet" flag for batch mode.
* Visor CMD: Fixed reading last command line in batch mode.
* Web Console: Added "Connected clusters" component in header.
* Web Console: Added duration and node ID in results header and 'Show query' modal.
* Web console: Fixed headers to avoid REST API caching in browsers (esp. IE).
* Web Console: Fixed resize of pinned columns.

GridGain In-Memory Data Fabric Professional Edition 2.1.8
------------------------------------
* Correct handling of half open communication connection.
* Discovery SPI: Ignite will skip non-Ignite servers during Discovery instead of failure
* Fixed an issue with infinitely deploying class on peer classloader.
* Fixed issue with affinity assignment mismatch after grid nodes restart.
* Fixed partition exchange hang due to SQL query running inside a transaction.
* Fixed peer classloader on SHARED mode
* Fixed UnsupportedOperationException on SSLSocket shutdown.
* Improved speed of discovery messages processing.
* Marshaller mappings and binary metadata are persisted to disk when registered locally
* ODBC: Added support of SQL_ATTR_CONNECTION_TIMEOUT
* ODBC: Added support of SQL_ATTR_QUERY_TIMEOUT
* ODBC: Driver now properly handles ungraceful TCP disconnects (keep-alive connection)
* ODBC: Fixed error that prevents from closing executed DML statement.
* SQL: Added support for Java 8 Date and Time API.
* Tx is properly rolled back on timeout.
* Visor Cmd: Fixed 'cache -a' in case of node filter.
* Web Console: Added "Copy to clipboard" action on Queries screen.
* Web Console: AngularJS 1.5.x upgraded to AngularJS 1.6.x.
* Web Console: Fixed memory leak on "Connection is not established" dialog.
* Web Console: Fixed Web Console under IE11.
* Web Console: Implemented progress indication for 'execute', 'scan' and 'export' actions on Queries screen.
* Web Console: Improved 'Export All' on Queries screen.
* Web Console: Select only server nodes for local queries on Queries screen.

GridGain In-Memory Data Fabric Professional Edition 2.1.7
------------------------------------
* Web Console: Fixed configuration generation for imported model in case of Oracle NUMBER(5) datatype.
* Web Console: Implemented configuration of services permissions.
* Fixed issue with blocking metadata read when called from discovery thread.
* SQL: Allowed multiple caches to share the same schema.
* Fixed @SpringResource injection via igniteSpringBean.
* Persisted marshaller mappings are checked during node startup, exception is thrown if corrupted mapping found.
* ODBC: Fixed SQLGetDiagRec behaviour on too small buffer size.
* Avoid infinite looping of communication and discovery reconnects.
* SQL Regex support in LINQ.

GridGain In-Memory Data Fabric Professional Edition 2.1.6
------------------------------------
* EstimatedRebalancingFinishTime and RebalancingStartTime metrics are exposed through MXBean interface.
* Fixed missed WAL REMOVE records for ATOMIC cache.
* Standalone WAL iterator handles WAL delete data records.
* Tx is properly rolled back on timeout.
* Fixed JVM crash on node deactivation.
* Fixed NPE when schemaIds for BinaryMetadata are null.
* Web Console: Optimized to run on big cluster with many caches.
* Web Console: Implemented cluster nodes log viewer.
* Web Console: Fixed output of big numbers in SQL query results.
* Web Console: Improved DDL support.
* Web console: Improved support for binary JDBC types on import from RDBMS.
* Visor CMD: Improved cache statistics.

GridGain In-Memory Data Fabric Professional Edition 2.1.5
------------------------------------
* Added ability to enable and disable rebalancing per-node.
* Added write throttling during an ongoing checkpoint to avoid zero performance drops.
* Added ability to convert WAL to human-redable form.
* Added ability to handle CacheObject from DataRecord in standalone WAL iterator.
* Added support for uninterruptable writes using AsyncFileIOFactory. Enabled with -DIGNITE_USE_ASYNC_FILE_IO_FACTORY=true.
* Improved comparing of inlined indexes.
* Local scans performance was improved up to 2 times in particular cases.
* DataStreamer performance improved.
* Default page size of durable memory is changed to 4KB.
* Default max memory size is changed from 80% to 20%.
* Introduced cache start/stop order on cluster activation.
* Web Console: Added sanitize user on save.
* Web Console: implemented refresh rate for Monitoring screen.
* Web Console: added charts with throughput and latency metrics for cache operations.
* Web Console: implemented cluster nodes log viewer.
* Web Console: Added version to footer.
* Web Console: Added "Lazy ResultSet" flag on Queries screen.
* Visor GUI: Fixed vulnerability of search in arbitrary folder.
* Visor CMD: Added missing configuration properties to "config" command.
* Visor CMD: Fixed script execution after alert throttling interval.
* ODBC: added SQLSTATE error codes support.
* ODBC: SQLBindParameter do not unbind parameter if the ParameterValuePtr is NULL anymore.
* ODBC: Fixed error on cursor closing before all the rows from the result fetched.
* ODBC: SQLRowCount now returns number of affected rows for non-batch queries.
* ODBC: Implemented SQL_ATTR_CONNECTION_DEAD connection attribute.
* ODBC: Implement SQLGetInfo for info types.
* ODBC: More detailed error message if the query parsing failed.
* JDBC Thin driver: added SQLSTATE error codes support.
* JDBC Thin driver: now returns correct JDBC, Database and Driver versions.
* JDBC Thin driver: added type conversions in accordance with specification.
* JDBC Thin driver: added SQLSTATE error codes support.
* JDBC Thin driver: added schema to connection string.
* JDBC Thin driver: added Statement.closeOnCompletion() support.
* JDBC: Metadata now expects "catalog" parameter match to empty.
* JDBC: added "enforceJoinOrder" flag to connection string.
* SQL: Fixed BigInteger data type handling.
* SQL: Added WRITE_SYNCHRONIZATION_MODE option to CREATE TABLE command.
* SQL: Added CREATE TABLE NOT NULL support.
* SQL: Added ALTER TABLE ADD COLUMN support.
* Optimized marshaller: fixed defaultWriteObject/defaultReadObject with NotActiveException.
* Infinite silent timeout while client with different configuration try enter topology.
* Reusing BinaryObjectBuilder can fails unexpectedly.
* Integration with BLAS via netlib.
* Improved tests for WAL iterator to check DataRecord entries.
* Excluded daemon nodes from snapshot operation topology.
* IgniteCache.loadCache() now doesn't block loading from store on remote nodes.
* Fixed consistent ID user in lightweight ClusterNode.
* Fixed offheap page evictions if loading not through data loader.
* Fixed overflow when using more than Short.MAX_VALUE partitions.
* Fixed exchange hang on partition eviction when partition is LOST.
* Fixed issue with client cache being created on server node.
* Fixed empty topology in discovery events when node has been segmented.
* Fixed race that could cause node hang-up during GridDhtTxFinishRequest processing.
* Fixed hanging on concurrent updates of binary metadata.
* Fixed PutAll for dependent objects.
* Fixed calculation of moving partitions count in partition map.
* Fixed transaction context clear.
* Fixed NullPointerException in JDBC Metadata getIndexInfo method.

GridGain In-Memory Data Fabric Professional Edition 2.1.4
------------------------------------
* Persistent store: implemented restore of marshaller mappings from file system on node startup
* Persistent store: added sorted and multithreaded modes in checkpointing
* Persistent store: added option to align pages in LFS partition files to pageSize
* Persistent store: added calculation of fill factor memory metric in persistent mode
* SQL: optimized memory consumption during query execution
* SQL: added QuerySqlField.inlineSize property
* SQL: added lazy query execution mode (SqlFieldsQuery.setLazy)
* SQL: added QueryEntity(Class keyClass, Class valueClass) constructor
* JDBC: added batch update support
* JDBC Thin Driver: added metadata support
* JDBC Thin Driver: added batch support
* Visor command-line: added possibility to not close connection in batch mode
* Visor command-line: added possibility to activate/deactivate cluster
* Web Console: added generation of QueryEntity.keyFields on model import from RDBMS
* Implemented thread context switching for optimistic transactions
* Replaced heap byte buffer with direct byte buffer in WAL records iterator
* Introduced more effective data structures to store and exchange partition counter values
* Added estimated time for rebalance start and completion to cache metrics
* Implemented optimization to process multiple discovery events in single partition map exchange
* Added API to add binary metadata locally
* Now validation error message are shown correctly for specific node
* Fixed issue with server nodes rejecting client connections on large topologies
* Fixed cluster active status on daemon node
* Fixed saving cache configuration on node join
* Fixed NPE in Lucene
* Fixed GridTimeoutProcessor hanging on node stop
* Fixed situation when partition counter may become 0 after exchange
* Fixed memory leak on page recovery
* Fixed storage of indexes with name length greater than 128
* Fixed AsynchronousCloseChannel in WAL
* Fixed partition files header alignment

GridGain In-Memory Data Fabric Professional Edition 2.1.3
---------------------------------------
* ODBC: added support for BINARY and VARBINARY types
* CPP: added support for BINARY and VARBINARY types
* CPP: Added Ignite::SetActive
* Persistent store: reduced time spent in TTL eviction during cache proxy leave
* Added an option to disable MBeans
* Fixed memory leak after scan query execution when expiry policy is used
* Fixed incorrect result returned by putIfAbsent when primary node fails
* Fixed TcpDiscoverySpi ignore maxMissedClientHeartbeats property
* Fixed asssertion in Write-Ahead Log during rollover
* Fixed stale sequence update for local partition map
* Fixed INFO log level escalation
* Decreased logging level for DataStreamer retries
* Added a proper exception when there are no available nodes for a partition during cache operations

GridGain In-Memory Data Fabric Professional Edition 2.1.2
---------------------------------------
* Machine learning: Implemented K-means clusterization algorithm optimized for distributed storages
* SQL: Improved LocalDate, LocalTime and LocalDateTime support for Java 8
* SQL: Added FieldsQueryCursor interface to get fields metadata for SqlFieldsQuery
* JDBC: Fixed a bug in PreparedStatement not clearing query parameters after execution
* ODBC: Implemented DML statement batching
* .NET: Added local collections joins support to LINQ
* CPP: Added Compute::Broadcast method
* Web Console: Added PK alias generation on Models screen
* Updated Lucene dependency to version 5.5.2
* Fixed possible deadlock in AtomicSequence when used inside transactions
* Fixed a bug when IS_EVICT_DISABLED flag was not cleared when cache store throws an exception
* Fixed striped pool starvation on unstable topology
* Fixed cluster instability on large near caches with metrics enabled
* Fixed a bug causing NullPointerException and hangs on certain unicode characters with index inlining

GridGain In-Memory Data Fabric Professional Edition 2.1.1
---------------------------------------
* Persistent cache store
* Added IgniteFuture.listenAsync() and IgniteFuture.chainAsync() mehtods
* Deprecated IgniteConfiguration.marshaller
* SQL: CREATE TABLE and DROP TABLE commands support
* SQL: New thin JDBC driver
* SQL: Improved performance of certain queries, when affinity node can be calculated in advance
* SQL: Fixed return type of AVG() function
* SQL: BLOB type support added to thick JDBC driver
* .NET: Automatic remote assembly loading
* .NET: NuGet-based standalone node deployment
* .NET: Added conditional data removeal via LINQ DeleteAll
* .NET: Added TimestampAttribute to control DateTime serialization mode
* CPP: Added Compute::call() method
* Web Console: Implemented support for UNIQUE indexes for key fields on import model from RDBMS
* Web Console: Added option to show full stack trace on Queries screen

GridGain In-Memory Data Fabric Professional Edition 2.0.1
---------------------------------------
* Introduced new page memory architecture
* Machine Learning beta: distributed algebra support for dense and sparse data sets
* Reworked and simplified API for asynchronous operations
* Custom thread pool executors for compute tasks
* Removed CLOCK mode in ATOMIC cache
* Deprecated schema-import utility in favor of Web Console
* Integration with Spring Data
* Integration with Hibernate 5
* Integration with RocketMQ
* Integration with ZeroMQ
* SQL: CREATE INDEX and DROP INDEX commands
* SQL: Ability to execute queries over specific set of partitions
* SQL: Improved REPLICATED cache support
* SQL: Updated H2 version to 1.4.195
* SQL: Improved performance of MIN/MAX aggregate functions
* ODBC: Added Time data type support
* .NET: Custom plugin API
* .NET: Generic cache store
* .NET: Binary types now can be registered dynamically
* .NET: LINQ: join, "contains" and DateTime property support
* CPP: Implemented Cache::Invoke
* CPP: Added remote filters support to continuous queries
* Web Console: Multi-cluster support
* Web Console: Possibility to configure Kubernetes IP finder
* Web Console: EnforceJoinOrder option on Queries screen
* Massive performance improvements

GridGain In-Memory Data Fabric Professional Edition 1.9.1-p1
------------------------------------
* Fixed failure when accessing BinaryObjectBuilder field value serialized with OptimizedMarshaller.
* Now back-pressure control supports for cache updates in PRIMARY_SYNC mode with backups.

GridGain In-Memory Data Fabric Professional Edition 1.9.1
------------------------------------
* Added Spark Shared RDD examples.
* Added pre-compiled benchmarks in binary fabric release.
* Added IP finder for Ignite nodes autodiscovery in Kubernetes.
* Bulk load of data into Ignite via JDBC is now possible with new JDBC driver mode.
* Improved aggregation, group by, offset/limit handling in subqueries.
* Fixed H2 license in LICENSE_FABRIC and enable Apache licenses to be listed in ignite-*-licenses.txt.
* IgniteRDD: Added support for Spark 2.0.
* Support TransactionScope API for Ignite transactions.
* SQL: query can utilize multiple threads.
* .NET: "ICollection.Contains" in LINQ support.
* .NET: DML API support.
* C++: Implemented LoadCache.
* C++: DML support.
* C++: ContinuousQuery support.
* Web Console: SQL screen redesigned for better usability.

GridGain In-Memory Data Fabric Professional Edition 1.8.5
------------------------------------
* Fix has been made to prevent GridQueryProcessor from using concrete classes of key or value in Binary mode when
  corresponding type is set to Object in query entity.
* Separate pool for services.
* Support of transactional updates in JDBC driver .
* Client node reconnects to cluster in case of communication problem instead of stopping.
* Fixed missed discovery data snapshot during exchange processing.
* Fix fail second client node join with continuous query and peer class loading enabled.
* Always use withKeepBinary flag for Jdbc driver when querying cache.

GridGain In-Memory Data Fabric Professional Edition 1.8.4
------------------------------------
* SQL: Sql query supports table alias.
* SQL: Add support for custom table name.

GridGain In-Memory Data Fabric Professional Edition 1.8.3
------------------------------------
* .NET: added DateTime properties support to LINQ provider.
* Web Console: agent is now able to connect through proxy.
* Hadoop: added automatic java.library.path property resolution.
* Fixed hang in Compute when IGNITE_BINARY_SORT_OBJECT_FIELDS system property is set to true.
* Fixed quite mode support for Slf4jLogger and JclLogger.

GridGain In-Memory Data Fabric Professional Edition 1.8.2
------------------------------------
* Hadoop: various performance improvements.
* TCP communication: implemented possibility to create separate incoming/outgoing connections.
* TCP communication: implemented possibility to create multiple connections between node.
* TCP communication: implemented NIO session balancing between NIO threads.
* Performance: added striped pool for cache message processing.

GridGain In-Memory Data Fabric Professional Edition 1.8.1
------------------------------------
* SQL: Added DML operations support (INSERT, UPDATE, DELETE, MERGE).
* SQL: Improved DISTINCT keyword handling in aggregates.
* Hadoop: Added MapR distribution support.
* Visor: Improved SQL statistics.
* .NET: ASP.NET session state store provider.
* .NET: Entity Framework second level cache.
* .NET: Custom loggers support: NLog, Apache log4Net.
* ODBC: Added DML operations support.
* ODBC: Added distributed joins support.
* ODBC: Added DSN support.
* ODBC: Performance improvements.
* Added Redis protocol support.
* Added transactions deadlock detection.
* Many stability and fault-tolerance fixes.

GridGain In-Memory Data Fabric Professional Edition 1.7.9-p1
------------------------------------
* Fixed ServiceProcessor hang while stopping on unstable topology.
* Client node reconnects to cluster in case of communication problem instead of stopping.

GridGain In-Memory Data Fabric Professional Edition 1.7.9
------------------------------------
* Fixed service redeploy on concurrent cancel and applying continuous query.

GridGain In-Memory Data Fabric Professional Edition 1.7.8
------------------------------------
* Allow to refresh client attributes during reconnect.
* ScanQuery performance optimization for on-heap cache.
* Improved error handling for SSL misconfiguration.

GridGain In-Memory Data Fabric Professional Edition 1.7.7
------------------------------------
* Fixed hang in Compute when IGNITE_BINARY_SORT_OBJECT_FIELDS system property is set to true.
* Fixed an issue with cache lock/unlock operations on changing topology.
* Optimized heap utilization in partition handling logic.
* Web Console: agent is now able to connect through proxy.

GridGain In-Memory Data Fabric Professional Edition 1.7.6
------------------------------------
* Fixed issue with IndexingSPI usage without default H2 Indexing enabled.
* Fixed issue with near cache synchronization when value is loaded from cache store.
* Fixed issue with ATOMIC cache update synchronization on changing topology.
* Fixed ScanQuery to properly handle pageSize property.
* SQL: implemented table alias support for SQL queries.
* SQL: implemented support for custom table name.
* Web Console: Added affinity configuration.
* Implemented dropping node from cluster if TcpCommunicationSpi fails to establish connection.

GridGain In-Memory Data Fabric Professional Edition 1.7.5
------------------------------------
* Added attribute based node filter.
* Added ability to pass objects to IndexingSpi in binary form.
* Fixed quite mode support for Slf4jLogger and JclLogger.
* Fixed an error causing exception when working with enums in binary form.

GridGain In-Memory Data Fabric Professional Edition 1.7.4
------------------------------------
* SQL: Fixed broken Local SQL fields query.
* SQL: Fixed "No query result found for request" exception which appeared sometimes during query execution under load.
* SQL: Implemented queries metrics for monitoring.
* Fixed bottleneck in partition exchange message processing.
* Fixed issues with offheap cache metrics.
* Fixed starvation in case when compute job calls cache.size().
* Preserved fields order in binary type metadata in accordance with schema.
* Client node does not start rest processor by default.
* Added possibility to specify timeout for service proxy.
* .NET: Added swap configuration.
* .NET: Added CacheConfiguration.ExpiryPolicyFactory.

GridGain In-Memory Data Fabric Professional Edition 1.7.3
------------------------------------
* .NET: Added NLog and Log4Net logger implementations.
* Web Console: Improved usability of SQL screen.
* Fixed bug in SSL handlers causing "Unsupported record version" exception.

GridGain In-Memory Data Fabric Professional Edition 1.7.2
------------------------------------
* IGFS: Performance improvements.
* Visor: Added reset cache metrics command.
* .NET: Added IClusterGroup.ForDaemons() method.
* .NET: Added ASP.NET Session-State Store Provider.
* Fixed issue with incorrect return value for cache invoke/getAndPut in transactional cache when primary owner leaves.

GridGain In-Memory Data Fabric Professional Edition 1.7.1
------------------------------------
* IGFS: Performance improvements.
* ODBC: Added distributed joins support.
* SQL: Updated H2 version from 1.3.175 to 1.4.191.

GridGain In-Memory Data Fabric Professional Edition 1.6.12
------------------------------------
* Fixed a problem with SqlFieldsQuery execution semantics in local mode.

GridGain In-Memory Data Fabric Professional Edition 1.6.11
------------------------------------
* Reduced size of discovery messages and amount of data stored in discovery and cache affinity history.
* Fixed bug in SSL handlers causing "Unsupported record version" exception.
* Task error output is now printed to the log instead of console.
* Fixed possible IgniteDataStreamer data loss on unstable topology.
* Backported 'copyOnRead' flag support for SQL queries.
* Improved binary objects raw byte array comparision.
* Fixed inconsistent fields ordering in binary objects.
* SQL: Added distinct aggregates support.
* SQL: Fixed "No query result found for request" exception which appeared sometimes during query execution under load.
* Hadoop: fixed secondary file system startup issue.
* CPP: Fixed compilation for GCC 4.4.7.
* ODBC: Created windows installer.

GridGain In-Memory Data Fabric Professional Edition 1.6.10
------------------------------------
* CPP: Fixed compilation for GCC 4.4.7.
* Hadoop: Added map-reduce plan debug output.
* Hadoop: Added ability to specify multiple cluster node address for job tracker.
* Hadoop: Added working directory support to IgniteHadoopIgfsSecondaryFileSystem.
* Fixed a bug in RoundRobinLoadBalancingSpi causing exception when running closures after client node reconnect.
* Fixed issue when transaction can hang if entry processor serialization failed.
* Fixed an issue in continuous queries backup queue handling for ATOMIC cache.
* Fixed a problem with SPI queries handling.
* Fixed messages loss under SSL.

GridGain In-Memory Data Fabric Professional Edition 1.6.9
------------------------------------
* Hadoop: Added MapR distribution support.
* JDBC POJO store: Implemented support for escaping of database schema, table and field names.
* Optimized internal cache threads usage: use single thread for expired entries clean-up instead of per-cache thread.
* Fixed a problem with message listener not being called under some conditions.
* Fixed executing internal tasks in case of custom LoadBalancingSPi.
* Fixed issue with lost discovery events after client reconnect.
* Fixed incorrect write of unregistered class names in BinaryMarshaller.

GridGain In-Memory Data Fabric Professional Edition 1.6.8
------------------------------------
* Implicit transactional invoke and getAndPut now returns correct result even when primary owner leaves.
* Fixed possible NullPointerException in continuous queries.
* Fixed passing incorrect old value to CacheInterceptor and ContinuousQuery during rebalancing.
* Fixed hang in FileSwapSpaceSpi when too large value is stored.
* Fixed StackOverflowError thrown from BinaryObject.toString().
* Performance and stability improvements.

GridGain In-Memory Data Fabric Professional Edition 1.6.7
------------------------------------
* ODBC: Added numeric, date/time and system functions support.
* ODBC: Added "socketSendBufferSize" and "socketReceiveBufferSize" configuration parameters.
* ODBC: Added DSN configuration dialog for Windows.
* Implemented deadlock detection support for optimistic transactions and near caches.
* Implemented resource injection support for cache EntryProcessors.
* Implemented transformers for cache queries.
* Fixed a problem with BinaryMarshaller handles resolution.
* CPP: Optimized SQL query cursors to better accommodate large result sets.

GridGain In-Memory Data Fabric Professional Edition 1.6.6
------------------------------------
* ODBC: Added string functions support.
* Automatic Persistence: Fixed issue with load large result set from MySQL.
* Import schema utility: fixed java code generation for group indexes.
* Visor: Added node id, node ip and values of metrics to alert log.
* Visor: Added output for off-heap metrics to "cache -a" command.

GridGain In-Memory Data Fabric Professional Edition 1.6.5
------------------------------------
* Decreased memory consumption in continuous queries.
* New methods for affinity-based execution in IgniteCompute.
* Fixed support for null values in SpringCacheManager.
* Fixed binary serialization of TreeSet and TreeMap classes.
* Fixed a bug in binary marshaller causing exceptions on a client after servers restart.
* IGFS: Added new secondary file system implementation based on local file system.
* IGFS: Performance improvements.

GridGain In-Memory Data Fabric Professional Edition 1.6.4
------------------------------------
* Fixed a bug in GridCacheTtlManager.
* Added basic implementation of address resolver.

GridGain In-Memory Data Fabric Professional Edition 1.6.3
------------------------------------
* .NET: Fixed a bug in IBinaryObjectBuilder causing type name to be lost.

GridGain In-Memory Data Fabric Professional Edition 1.6.2
------------------------------------
* Fixed CountDownLatch recreation.
* CPP: Fixed string deserialization performance issue.
* .NET: Added NuGet distribution.

GridGain In-Memory Data Fabric Professional Edition 1.6.1
------------------------------------
Ignite .NET:
* Added LINQ Provider for cache SQL queries
* Added native configuration mechanism (C#, app.config, web.config - instead of Spring XML)
* Added NuGet distribution
* Added AtomicSequence and AtomicReference data structures
* Binaries are now AnyCPU (instead of separate x64/x86)
* Java-based services can be called the same way as .NET service are called
* Added java-based continuous query filters
* Added automatic Java detection: no need for JAVA_HOME
* Added ability to use Java-based remote filters in continuous queries.

Ignite C++:
* Added Transactions API for C++ client.
* Added Date and Timestamp types implementation for C++ client.
* Simplified Autotools build process for C++ client.

Ignite:
* Added Cache Store implementation backed by Cassandra DB.
* Added ODBC driver for Ignite.
* Added Web Console for Ignite.
* ODBC: Added decimal type support.

Changes in GridGain In-Memory Data Fabric Professional Edition 1.5.33
------------------------------------
* No changes.

Changes in GridGain In-Memory Data Fabric Professional Edition 1.5.32
------------------------------------
* Fixed SSL context initialization for REST server.

Changes in GridGain In-Memory Data Fabric Professional Edition 1.5.31
------------------------------------
* Improved local Ignite instance handling during serialization and deserialization.

Changes in GridGain In-Memory Data Fabric Professional Edition 1.5.30
------------------------------------
* Ignite Cache: new API that lets calculating size per particular partition.
* Visor CMD: Added skip confirm in batch mode. Improved kill command.
* Hadoop: Implemented new weight-based map-reduce planner (IgniteHadoopWeightedMapReducePlanner).
* .NET: Added ability to implement custom affintiy functions.
* Added support of EXPIRED events to continuous queries.
* Data Streamer: ability to set timeout for a number of methods.
* Fixed NPE in Ignite Services internals that happened when a service was executed before system cache was started.
* Fixed NPE in OptimizedMarshaller when pool of streams is used.
* Fixed the issue when it wasn't possible to deserialize an object on Java side if the object was placed in the cache on C++ or .Net side.
* Fixed SQL indexes processing under Java 8.

GridGain In-Memory Data Fabric Professional Edition 1.5.29
------------------------------------
* Binary objects: performance improvements.
* IGFS: performance and usability improvements.
* Fixed issue with unnecessary discovery messages resend that may cause slow discovery message delivery.
* Fixed possible IO message loss during reconnect.
* Additional system properties for troubleshooting logging.
* Fixed issue when job ID was generated in a non unique way.

GridGain In-Memory Data Fabric Professional Edition 1.5.28
------------------------------------
* IGFS: Usability improvements.
* Fixed performance degradation when a cache is preloaded from many IgniteDataStreamers in parallel from the same node.
* Fixed issue with service redeployment after topology change.

GridGain In-Memory Data Fabric Professional Edition 1.5.27
------------------------------------
* IGFS: Fixed a problem with native libraries loading.
* REST HTTP: Updated from json-lib 2.4 to Jackson 2.7.5.
* External addresses are now registered in IP finder.
* Client node's addresses are now not registered in IP finder.

GridGain In-Memory Data Fabric Professional Edition 1.5.26
------------------------------------
* IGFS: Performance improvements.
* Spark: added IgniteRDD.withKeepBinary() method for full support of binary objects.
* Added Cassandra-based persistent store implementation.
* Continuous queries: warning will be printed on client nodes instead of exception trace if the remote filter is not in classpath.
* Fixed missing IgniteCountdownLatch updates that can lead to a deadlock or wrong behavior.
* Fixed local store conflict resolving with keepBinary flag set to false.
* Fixed addresses duplication at IP finder.
* ZookeeperIpFinder now properly handles reconnects.

GridGain In-Memory Data Fabric Professional Edition 1.5.25
------------------------------------
* IGFS: Performance improvements for client mode.
* Hadoop: fixed inefficient memory usage during mapper output processing.
* Fixed marshalling of arrays of primitives types when their are being placed into a cache.
* Spark: added transformer closure to savePairs(..) and saveValues(..) methods of IgniteRDD.
* Spark: fixed support for java.sql.Date in queries executed from IgniteRDD.

GridGain In-Memory Data Fabric Professional Edition 1.5.24
------------------------------------
* IGFS: Added user name mapper to properly deal with security in complex Hadoop environments.
* Validating that the same IgniteConfiguration instance is not used more than once.
* Fixed marshalling of Void type with Binary marshaller.
* Fixed metrics calculation for OFFHEAP_TIRED mode.
* ScanQuery and localEntries now honor keepBinary flag.
* Zookeeper module dependencies set fixed.

GridGain In-Memory Data Fabric Professional Edition 1.5.23
------------------------------------
* IGFS: Improved error logging for operations in DUAL mode.
* Added Ignition.getOrStart method.
* Spark: support BigDecimal for queries executed from IgniteRDD.
* Mesos integration: added possibility to configure http server timeout.
* Fixed a bug with Ignite API usage during node stop.
* Fixed a bug with 'keepBinary' handling for CacheInterceptor.
* Fixed a bug with CacheEntryEvent cleanup for filtered events.
* Fixed a bug with broken synchronization between cache entry expiration and update.

GridGain In-Memory Data Fabric Professional Edition 1.5.22
------------------------------------
* Fixed UTF-16 surrogate pairs marshalling.
* Fixed local store behavior at cross cache transactions.
* Fixed memory leak in IgniteH2Indexing.
* Added backup partitions storing to local store default behavior.
* ScanQueries over local partitions performance optimisation.
* Visor CMD: Added ability to attach custom scripts to alerts.
* WebSession: fixed a bug causing invalidated session to be returned from HttpServletRequest.getSession() call.
* Hadoop: fixed a bug causing exception during job resource unpacking.

GridGain In-Memory Data Fabric Professional Edition 1.5.21
------------------------------------
* Web sessions: minor fix to prevent NullPointerException in some special case.

GridGain In-Memory Data Fabric Professional Edition 1.5.20
------------------------------------
* Web sessions: user session classes are no longer needed on server nodes.

GridGain In-Memory Data Fabric Professional Edition 1.5.19
------------------------------------
* Added cache deadlock detection.
* Fixed a bug causing object deserialization when local store is configured for cache.
* Fixed a problem with incorrect classloader picked in OSGI environment.
* Hadoop: fixed a bug causing exception during MR planning when input split file doesn't exist.

GridGain In-Memory Data Fabric Professional Edition 1.5.18
------------------------------------
* Fixed potential thread starvation during cache rebalancing.
* Fixed deadlock in services deployment when Spring is used.
* Mesos integration: added possibility to load Ignite package from file system.

GridGain In-Memory Data Fabric Professional Edition 1.5.17
------------------------------------
* Performance optimization for transactions that do not acquire locks.
* Optimized memory usage on server nodes to allow more client connections.
* Added support for join timeout while registering local addresses with IP finder in TcpDiscoverySpi.

GridGain In-Memory Data Fabric Professional Edition 1.5.16
------------------------------------
* Added asynchronous execution of ContinuousQuery's remote filter and local listener.

GridGain In-Memory Data Fabric Professional Edition 1.5.15
------------------------------------
* Fixed discovery and node start procedure issues that may cause slowdowns during larger topologies start (>100 nodes).

GridGain In-Memory Data Fabric Professional Edition 1.5.14
------------------------------------
* Fixed a race condition when evicted offheap data was not delivered to query engine leading to query timeout.

GridGain In-Memory Data Fabric Professional Edition 1.5.12
------------------------------------
* Improved marshaller behavior during cluster topology change.
* .NET: Added ability to use Java-based remote filters in continuous queries.

GridGain In-Memory Data Fabric Professional Edition 1.5.11
------------------------------------
* Fixed cluster stability with 500+ clients.
* Fixed background cache partition map exchange not to flood network.
* Fixed BinaryObjectOffHeapImpl leak to public code.
* Fixed BinaryContext to honor custom loader set through IgniteConfiguration.
* Fixed continuous query not to call listener with empty set of events.
* IGFS: Massive performance and stability fixes.
* Many stability and fault-tolerance fixes.

GridGain In-Memory Data Fabric Professional Edition 1.5.10
------------------------------------
* Fixed continuous query listener notification handling.

GridGain In-Memory Data Fabric Community Edition 1.5.9
---------------------------------------
* Fixed issue with AWS dependencies.
* Fixed excessive thread stack usage in case of high contention on cache entries.
* Fixed continuous queries to send filter factory instead of filter.
* Fixed NPE during rebalancing.
* Fixed system caches not to use user-defined TransactionConfiguration.
* Fixed GridClosureProcessor internal closures to be deserialized by BinaryMarshaller.
* Fixed web session clustering with WebLogic.
* Fixed value copying in entry processor with OptimizedMarshaller.
* Fixed continuous query listener notification during concurrent key put.
* Fixed service proxy not to make remote call for methods declared in java.lang.Object.
* Added support for JTA transactions via synchronization callback.

GridGain In-Memory Data Fabric Community Edition 1.5.8
---------------------------------------
* Fixed java proxies deserialization with optimized marshaller.
* Fixed excessive thread stack usage in case of high contention on cache entries.
* Fixed SKIP_STORE flag behavior for transaction entry.
* Fixed thread safety for TcpDiscoveryMulticastIpFinder.
* IGFS: Fixed file properties when running over secondary file system.
* IGFS: Fixed issues with Kerberized HDFS.

GridGain In-Memory Data Fabric Community Edition 1.5.7
---------------------------------------
* Added method to get versioned cache entry.
* Added ability to get partition 'updateCntr' with continuous query public API.
* Fixed several ClassNotFoundException in OSGi environment.
* Fixed continuous query notification with single node topology.
* Fixed continuous query notification with cache listeners.
* Fixed continuous query deployment in case originating node has left.
* Fixed NPE in GridMergeIndex.
* Fixed race condition on load cache on changing topology.
* Fixed OOME when OFFHEAP_TIERED mode is used.
* Fixed support for classes with equal simple name for binary marshaller.

GridGain In-Memory Data Fabric Community Edition 1.5.6
---------------------------------------
* Support of package-private implementations for Ignite Services.
* Fixed unintentional deserialization of BinaryObjects in OFFHEAP mode with peer class loading enabled.

GridGain In-Memory Data Fabric Community Edition 1.5.5
---------------------------------------
* Fixed the issue when CacheStore was updated even if EntryProcessor didn't update an entry.
* Fixed execution of Splunk MapReduce jobs on top of Ignite MR.

GridGain In-Memory Data Fabric Community Edition 1.5.4
---------------------------------------
* Passed update notifier flag to discovery data to be sure that all cluster nodes have the same notifier status as the first node.
* Optimized cache 'get' operations on affinity node.
* Supported Ignite Compute tasks cancellation for Ignite.NET.
* Fixed race in marshalling logic that could arise when several nodes are running on the same host.
* Fixed marshalling of Java collection and maps for BinaryMarshaller.
* Fixed eviction policy notification if swap or off heap is enabled and an entry is not loaded during preloading.
* Fixed redeployment issues happened with ScanQueries filters.
* Fixed an issue in ContinuousQueries that caused missing of notifications.

GridGain In-Memory Data Fabric Community Edition 1.5.3
---------------------------------------
* IGFS: Added pluggable factory interface for Hadoop FileSystem creation.
* IGFS: Added configuration flag to disable default path modes under "/ignite" folder.
* IGFS: Various stability fixes.

GridGain In-Memory Data Fabric Community Edition 1.5.1
---------------------------------------
* Ignite.NET: Initial Release.
* Ignite C++: Initial Release.
* Massive performance improvements for cache operations and SQL.
* Added new binary cache object marshalling implementation.
* Added IgniteSemaphore data structure.
* Added MQTT Streamer.
* Added Twitter Streamer.
* Added Ignite Sink (integration with Apache Flume).
* Added Apache Camel Streamer.
* Added OSGi support.
* Fixed optimistic serializable transactions: implemented "deadlock-free" locking algorithm.
* Fixed failover for continuous queries.
* Fixed compilation and runtime errors under OpenJDK and IBM JDK.
* Fixed Integer.size limitation for cache.
* Fixed and improved cache types configuration.
* Fixed cache rebalancing.
* Many stability and fault-tolerance fixes.
* Added example to demonstrate the usage of EntryProcessor.

Complete list of closed issues: https://issues.apache.org/jira/issues/?jql=project%20%3D%20IGNITE%20AND%20fixVersion%20%3D%201.5%20AND%20status%20%3D%20closed

GridGain In-Memory Data Fabric Community Edition 1.4.1
---------------------------------------
* Added SSL support to communication and discovery.
* Added support for log4j2.
* Added versioned entry to cache API.
* Added client node based JDBC driver implementation.
* Fixed IGNITE_HOME resolution with JBoss.
* Fixed configuration file resolution on classpath.
* Fixed MBean registration for caches.
* Fixed near cache evictions with offheap.
* Fixed SQL queries group index logic.
* Fixed memory leak in partitioned transactional cache.
* Fixed auto-retries for cache operations in recoverable cases.
* Fixed affinity routing in compute grid.
* Many stability and fault-tolerance fixes.

GridGain In-Memory Data Fabric Community Edition 1.3.3
---------------------------------------

* Added SSL support to communication and discovery.
* Fixed IGNITE_HOME resolution with JBoss.
* Fixed configuration file resolution on classpath.
* Fixed MBean registration for caches.
* Many stability and fault-tolerance fixes.

GridGain In-Memory Data Fabric Community Edition 1.3.1
---------------------------------------

* Added auto-retries for cache operations in recoverable cases.
* Added integration with Apache YARN.
* Fixed several issues with JTA integration.
* Fixed several issues with Hibernate L2 cache.
* Stability fixes for TCP discovery SPI.
* Stability fixes for onheap and offheap SQL queries.
* Bug fixes in In-Memory Accelerator For Apache Hadoop.
* Many stability and fault-tolerance fixes.

GridGain In-Memory Data Fabric Community Edition 1.1.4
---------------------------------------

* Added dropping of slow clients.
* Fixed several connectivity issues.

GridGain In-Memory Data Fabric Community Edition 1.1.3
---------------------------------------

* Fixed cache store configuration validation.

GridGain In-Memory Data Fabric Community Edition 1.1.2
---------------------------------------

* Added client mode to TCP discovery SPI.
* Added memory based evictions.
* Added integration with Apache Spark.
* Added integration with Apache Mesos.
* Added IgniteCache.getAllOutTx() method.
* Added serializable cache store factories for built-in stores.
* Bug fixes in In-Memory Accelerator For Apache Hadoop.
* Many stability and fault-tolerance fixes.

GridGain In-Memory Data Fabric Community Edition 1.0.5
---------------------------------------

* Fixed unnecessary value sending for near cache with near lock response
* Optimized getOrCreateCache methods for case when cache is already started locally

GridGain In-Memory Data Fabric Community Edition 1.0.4
---------------------------------------

* Added Google Compute Engine TCP discovery IP finder.
* Added generic cloud TCP discovery IP finder (based on jclouds).
* Added SortedEvictionPolicy.
* Added chaining for IgniteConfiguration and CacheConfiguration setters.
* Added AffinityUuid class for easier generation of unique collocated keys.
* Added support for cache configuration templates.
* Added support for (*) star notation in cache names.
* Added "collocated" mode for SQL queries.
* Improved expiry policy handling (TTL based evictions) to avoid thrashing.
* Fixed job continuations.
* Fixed compilation and runtime with OpenJDK 7 & 8
* Fixed SQL Union support
* Fixed Word-Count streaming example to produce better results.
* Removed edtFTPj scanner from URI deployment due to licensing issues.
* Made deployment scanners for URI-based deployment pluggable.
* Many stability and fault-tolerance fixes.

GridGain In-Memory Data Fabric Community Edition 1.0
---------------------------------------

* Simplified query API.
* Added automatic aggregation, grouping, and sorting support to SQL queries.
* Added dynamic caches, i.e. ability to start and stop caches during runtime.
* Changed IgniteDataLoader to IgniteDataStreamer.
* Added automatic transformation support to streaming.
* Removed old streaming APIs in favor of new IgniteDataStreamer.
* Added many examples using Java8 lambdas and streams.
* Added new streaming examples based on new streaming and SQL querying APIs.
* Added automatic schema-import demo under examples.
* Improved javadoc documentation.
* Added ability to automatically exclude LGPL optional dependencies during build.


GridGain In-Memory Data Fabric Community Edition 1.0 RC3
-------------------------------------------

This is the first release of Apache Ignite project. The source code in large part is based
on the 7 year old GridGain In-Memory Data Fabric, open source edition, v. 6.6.2, which was
donated to Apache Software Foundation in September 2014.

The main feature set of Ignite In-Memory Data Fabric includes:
* Advanced Clustering
* Compute Grid
* Data Grid
* Service Grid
* IGFS - Ignite File System
* Distributed Data Structures
* Distributed Messaging
* Distributed Events
* Streaming & CEP
