# Kafka Streams Memory Management for Confluent Platform Kafka Streams memory management controls how much RAM each application instance uses for internal *record caches*, which buffer and compact records before they are written to state stores or forwarded to downstream processors. You configure the total cache size with the `statestore.cache.max.bytes` parameter, and Kafka Streams splits this budget evenly across the application’s stream threads. The record caches are implemented slightly differently in the DSL and Processor API. ## Record caches in the DSL You can specify the total memory (RAM) size of the record cache for an instance of the processing topology. The following `KTable` instances leverage it: * Source `KTable`: `KTable` instances that are created via `StreamsBuilder#table()` or `StreamsBuilder#globalTable()`. * Aggregation `KTable`: instances of `KTable` that are created as a result of [aggregations](dsl-api.md#streams-developer-guide-dsl-aggregating). For such `KTable` instances, the record cache is used for: * Internal caching and compacting of output records before they are written by the underlying stateful [processor node](../concepts.md#streams-concepts-processor) to its internal state stores. * Internal caching and compacting of output records before they are forwarded from the underlying stateful [processor node](../concepts.md#streams-concepts-processor) to any of its downstream processor nodes. Use the following example to understand the behaviors with and without record caching. In this example, the input is a `KStream` with the records `: , , , `. The focus in this example is on the records with key == `A`. * An [aggregation](dsl-api.md#streams-developer-guide-dsl-aggregating) computes the sum of record values, grouped by key, for the input and returns a `KTable`. * **Without caching**: a sequence of output records is emitted for key `A` that represent changes in the resulting aggregation table. The parentheses (`()`) denote changes, the left number is the new aggregate value and the right number is the old aggregate value: `, , `. * **With caching**: a single output record is emitted for key `A` that would likely be compacted in the cache, leading to a single output record of ``. This record is written to the aggregation’s internal state store and forwarded to any downstream operations. The cache size is specified by using the `statestore.cache.max.bytes` parameter, which is a global setting per processing topology: ```java // Enable record cache of size 10 MB. Properties streamsConfiguration = new Properties(); streamsConfiguration.put(StreamsConfig.STATESTORE_CACHE_MAX_BYTES_CONFIG, 10 * 1024 * 1024L); ``` This parameter controls the number of bytes allocated for caching. Specifically, for a processor topology instance with `T` threads and `C` bytes allocated for caching, each thread has an even `C/T` bytes to construct its own cache and use as it sees fit among its tasks. This means that there are as many caches as there are threads, but no sharing of caches across threads happens. The basic API for the cache is made of `put()` and `get()` calls. Records are evicted using a simple LRU scheme after the cache size is reached. The first time a keyed record `R1 = ` finishes processing at a node, it is marked as dirty in the cache. Any other keyed record `R2 = ` with the same key `K1` that is processed on that node during that time will overwrite ``, this is referred to as “being compacted”. This has the same effect as [Kafka’s log compaction](/kafka/design/log_compaction.html), but happens earlier, while the records are still in memory, and within your client-side application, rather than on the server-side (that is, the Apache Kafka® broker). After flushing, `R2` is forwarded to the next processing node and then written to the local state store. The semantics of caching are that data is flushed to the state store and forwarded to the next downstream processor node whenever the earliest of `commit.interval.ms` or `statestore.cache.max.bytes` (cache pressure) hits. Both `commit.interval.ms` and `statestore.cache.max.bytes` are global parameters. As such, it is not possible to specify different parameters for individual nodes. Here are example settings for both parameters based on desired scenarios. - To turn off caching the cache size can be set to zero: ```java // Disable record cache Properties streamsConfiguration = new Properties(); streamsConfiguration.put(StreamsConfig.STATESTORE_CACHE_MAX_BYTES_CONFIG, 0); ``` With default settings, caching is enabled in Kafka Streams and RocksDB. - To enable caching but still have an upper bound on how long records are cached, you can set the commit interval. In this example, it is set to 1000 milliseconds: ```java Properties streamsConfiguration = new Properties(); // Enable record cache of size 10 MB. streamsConfiguration.put(StreamsConfig.STATESTORE_CACHE_MAX_BYTES_CONFIG, 10 * 1024 * 1024L); // Set commit interval to 1 second. streamsConfiguration.put(StreamsConfig.COMMIT_INTERVAL_MS_CONFIG, 1000); ``` The effect of these two configurations is described in the figure below. The records are shown using four keys: blue, red, yellow, and green. Assume the cache has space for only three keys. - When the cache is disabled (a), all of the input records are output. - When the cache is enabled (b): - Most records are output at the end of commit intervals (for example, at `t1` a single blue record is output, which is the final over-write of the blue key up to that time). - Some records are output because of cache pressure (that is, before the end of a commit interval). For example, see the red record before `t2`. With smaller cache sizes we expect cache pressure to be the primary factor that dictates when records are output. With large cache sizes, the commit interval is the primary factor. - The total number of records output has been reduced from 15 to 8.  ## Record caches in the Processor API You can specify the total memory (RAM) size of the record cache for an instance of the processing topology. It is used for internal caching and compacting of output records before they are written from a stateful processor node to its state stores. The record cache in the Processor API does not cache or compact any output records that are being forwarded downstream. This means that all downstream processor nodes can see all records, whereas the state stores see a reduced number of records. This does not impact correctness of the system, but is a performance optimization for the state stores. For example, with the Processor API you can store a record in a state store while forwarding a different value downstream. Following from the example first shown in section [State stores](processor-api.md#streams-developer-guide-state-store), to enable caching, you can add the `withCachingEnabled` call. ```java StoreBuilder countStoreBuilder = Stores.keyValueStoreBuilder( Stores.persistentKeyValueStore("Counts"), Serdes.String(), Serdes.Long()) .withCachingEnabled() ``` ## RocksDB Each instance of RocksDB allocates off-heap memory for a block cache, index and filter blocks, and memtable (write buffer). Critical configs include `block_cache_size`, `write_buffer_size` and `max_write_buffer_number`. These can be specified through the `rocksdb.config.setter` configuration. The Kafka Streams record cache and the RocksDB cache are not mutually exclusive. By default, both are enabled in a Kafka Streams app. The record cache (on heap) is particularly useful for optimizing writes by reducing the number of updates to local state and changelog topics. The RocksDB block cache (off heap) optimizes reads. Confluent recommends changing RocksDB’s default memory allocator, because the default allocator may lead to increased memory consumption. To change the memory allocator to `jemalloc`, set the `LD_PRELOAD` environment variable before you start your Kafka Streams application: ```shell # example: install jemalloc (on Debian) apt install -y libjemalloc-dev # set LD_PRELOAD before you start your Kafka Streams application export LD_PRELOAD="/usr/lib/x86_64-linux-gnu/libjemalloc.so" ``` The RocksDB block cache only serves reads and doesn’t soak up writes. Writes always go into the memtable. Write traffic to disk is determined by the memtable being flushed, which happens during a commit or when the memtables are “full”. By default, you need three 16 MB memtables to fill up before flushing. You may be able to determine the right size by leveraging the RocksDB statistics to determine which limit you hit first: the commit interval or the memtable size limit. The default RocksDB block-cache size is 50 MB per store, but the default size of the Kafka Streams record cache is 10 MB for caching for the entire instance. If you have a large number of stores, this 50 MB default can be too high. The memory usage across all instances can be bounded, limiting the total off-heap memory of your Kafka Streams application. To do so you must configure RocksDB to cache the index and filter blocks in the block cache, limit the memtable memory through a shared [WriteBufferManager](https://github.com/facebook/rocksdb/wiki/Write-Buffer-Manager) and count its memory against the block cache, and then pass the same `Cache` object to each instance. See [RocksDB Memory Usage](https://github.com/facebook/rocksdb/wiki/Memory-usage-in-RocksDB) for details. An example `RocksDBConfigSetter` implementing this is shown below: ```java public static class BoundedMemoryRocksDBConfig implements RocksDBConfigSetter { static { RocksDB.loadLibrary(); } // See #1 below private static org.rocksdb.Cache cache = new org.rocksdb.LRUCache(TOTAL_OFF_HEAP_MEMORY, -1, false, INDEX_FILTER_BLOCK_RATIO); private static org.rocksdb.WriteBufferManager writeBufferManager = new org.rocksdb.WriteBufferManager(TOTAL_MEMTABLE_MEMORY, cache); @Override public void setConfig(final String storeName, final Options options, final Map configs) { BlockBasedTableConfig tableConfig = (BlockBasedTableConfig) options.tableFormatConfig(); // These three options in combination will limit the memory used by RocksDB to the size passed to the block cache (TOTAL_OFF_HEAP_MEMORY) tableConfig.setBlockCache(cache); tableConfig.setCacheIndexAndFilterBlocks(true); options.setWriteBufferManager(writeBufferManager); // These options are recommended to be set when bounding the total memory // See #2 below tableConfig.setCacheIndexAndFilterBlocksWithHighPriority(true); tableConfig.setPinTopLevelIndexAndFilter(true); // See #3 below tableConfig.setBlockSize(BLOCK_SIZE); options.setMaxWriteBufferNumber(N_MEMTABLES); options.setWriteBufferSize(MEMTABLE_SIZE); // Enable compression (optional). Compression can decrease the required storage // and increase the CPU usage of the machine. For CompressionType values, see // https://javadoc.io/static/org.rocksdb/rocksdbjni/6.4.6/org/rocksdb/CompressionType.html. options.setCompressionType(CompressionType.LZ4_COMPRESSION); options.setTableFormatConfig(tableConfig); } @Override public void close(final String storeName, final Options options) { // Cache and WriteBufferManager should not be closed here, as the same objects are shared by every store instance. } } ``` Footnotes: : 1. `INDEX_FILTER_BLOCK_RATIO` can be used to set a fraction of the block cache to set aside for “high priority” (aka index and filter) blocks, preventing them from being evicted by data blocks. The boolean parameter in the cache constructor lets you control whether the cache should enforce a strict memory limit by failing the read or iteration in the rare cases where it might go larger than its capacity. For more information, see the full signature of the [LRUCache constructor](https://github.com/facebook/rocksdb/blob/master/java/src/main/java/org/rocksdb/LRUCache.java#L72). 2. This must be set in order for `INDEX_FILTER_BLOCK_RATIO` to take effect (see footnote 1) as described in the [RocksDB docs](https://github.com/facebook/rocksdb/wiki/Block-Cache#caching-index-and-filter-blocks). 3. You might want to modify the default [block size](https://github.com/apache/kafka/blob/trunk/streams/src/main/java/org/apache/kafka/streams/state/internals/RocksDBStore.java) per these instructions from the [RocksDB GitHub](https://github.com/facebook/rocksdb/wiki/Memory-usage-in-RocksDB#indexes-and-filter-blocks). A larger block size means index blocks will be smaller, but the cached data blocks might contain more cold data that would otherwise be evicted. While Confluent recommends setting at least the preceding configurations, the specific options that yield the best performance are workload dependent, and you should consider experimenting with these to determine the best choices for your specific use case. Keep in mind that the optimal configurations for one application might not apply to one with a different topology or input topic. In addition to the preceding recommended configurations, you may want to consider using partitioned index filters as described by the [RocksDB docs](https://github.com/facebook/rocksdb/wiki/Partitioned-Index-Filters). ## Other memory usage There are other modules inside Kafka that allocate memory during runtime. They include the following: * Producer buffering, managed by the producer config `buffer.memory`. * Consumer buffering, currently not strictly managed, but can be indirectly controlled by fetch size, i.e., `fetch.max.bytes` and `fetch.max.wait.ms`. * Both producer and consumer also have separate TCP send / receive buffers that are not counted as the buffering memory. These are controlled by the `send.buffer.bytes` / `receive.buffer.bytes` configs. * Deserialized objects buffering: after `consumer.poll()` returns records, they are deserialized to extract timestamp and buffered in the streams space. This was indirectly controlled by the `buffered.records.per.partition` config, which is deprecated starting in Confluent Platform 7.4. #### NOTE This website includes content developed at the [Apache Software Foundation](https://www.apache.org/) under the terms of the [Apache License v2](https://www.apache.org/licenses/LICENSE-2.0.html).