[[["์ดํดํ๊ธฐ ์ฌ์","easyToUnderstand","thumb-up"],["๋ฌธ์ ๊ฐ ํด๊ฒฐ๋จ","solvedMyProblem","thumb-up"],["๊ธฐํ","otherUp","thumb-up"]],[["์ดํดํ๊ธฐ ์ด๋ ค์","hardToUnderstand","thumb-down"],["์๋ชป๋ ์ ๋ณด ๋๋ ์ํ ์ฝ๋","incorrectInformationOrSampleCode","thumb-down"],["ํ์ํ ์ ๋ณด/์ํ์ด ์์","missingTheInformationSamplesINeed","thumb-down"],["๋ฒ์ญ ๋ฌธ์ ","translationIssue","thumb-down"],["๊ธฐํ","otherDown","thumb-down"]],["์ต์ข ์ ๋ฐ์ดํธ: 2025-09-03(UTC)"],[[["\u003cp\u003eAlloyDB Omni employs adaptive algorithms to manage memory, dynamically adjusting the shared buffer size for optimal performance.\u003c/p\u003e\n"],["\u003cp\u003eThe default \u003ccode\u003eshared_buffers\u003c/code\u003e parameter is set to \u003ccode\u003e0\u003c/code\u003e, which automatically allocates up to 80% of system memory for the shared buffer cache, starting initially at 10% of that limit.\u003c/p\u003e\n"],["\u003cp\u003eUsers can customize the \u003ccode\u003eshared_buffers\u003c/code\u003e upper limit by modifying the \u003ccode\u003epostgresql.conf\u003c/code\u003e file, and AlloyDB Omni will use this custom value as the maximum shared buffer size and start with this specified size.\u003c/p\u003e\n"],["\u003cp\u003eAlloyDB Omni can either shrink or grow the shared buffer backing up size at runtime, and when the columnar engine is enabled, the dynamic shared buffer size is determined after subtracting the memory used by the columnar engine from the 80% of total available memory or cgroup.\u003c/p\u003e\n"],["\u003cp\u003eAlloyDB Omni manages huge pages for improved performance, falling back to transparent huge pages (THP) or 4k pages if huge pages are unsupported, with warnings in the container logs if setup is needed.\u003c/p\u003e\n"]]],[],null,["# About automatic memory management\n\nSelect a documentation version: Current (16.8.0)keyboard_arrow_down\n\n- [Current (16.8.0)](/alloydb/omni/current/docs/automatic-memory-management)\n- [16.8.0](/alloydb/omni/16.8.0/docs/automatic-memory-management)\n- [16.3.0](/alloydb/omni/16.3.0/docs/automatic-memory-management)\n- [15.12.0](/alloydb/omni/15.12.0/docs/automatic-memory-management)\n- [15.7.1](/alloydb/omni/15.7.1/docs/automatic-memory-management)\n- [15.7.0](/alloydb/omni/15.7.0/docs/automatic-memory-management)\n- [15.5.5](/alloydb/omni/15.5.5/docs/automatic-memory-management)\n- [15.5.4](/alloydb/omni/15.5.4/docs/automatic-memory-management)\n\n\u003cbr /\u003e\n\nAlloyDB Omni uses adaptive algorithms for memory management.\n\n\u003cbr /\u003e\n\nYou can decide the upper limit of the shared buffer on starting AlloyDB Omni. If you don't set the upper limit, AlloyDB Omni automatically sets the shared buffer backing up size to 80% of system memory. The initial backing up size of the shared buffer can be different from the upper limit.\n\nAlloyDB Omni consists of an intelligent memory worker which constantly monitors the memory status, and tunes shared buffer backing up size for best performance when caching data.\n\nBy default the `shared_buffers` parameter is set to `0` which is a special value that sets the upper limit of the size of the `shared buffers` cache to 80% of system memory. AlloyDB Omni starts at 10% of the `shared_buffers` upper limit. If `shared_buffers` is overridden by a custom value, then AlloyDB Omni respects the value as the upper limit of the `shared_buffers` size, and starts with that specified custom size.\n| **Note:** Memory worker can still decide to shrink, and later grow shared buffer backing up size at runtime.\n\nTo specify a custom size, edit the `postgresql.conf` configuration file. For example, you can set `shared_buffers` to `1GB` using either of the following:\n\n- `docker run --name `\u003cvar translate=\"no\"\u003eCONTAINER_NAME\u003c/var\u003e` -e INITDB_ARGS=\"-c shared_buffers=1GB\" $image`\n\n- `docker run --name `\u003cvar translate=\"no\"\u003eCONTAINER_NAME\u003c/var\u003e` $image -c shared_buffers=1GB`\n\n Replace \u003cvar translate=\"no\"\u003eCONTAINER_NAME\u003c/var\u003e with the name that you assigned to the AlloyDB Omni container when you installed it.\n\nOptimize query performance\n--------------------------\n\nThe default value of the `shared_buffers` parameter works for common scenarios.\n\nHowever, you can tune the value for best performance.\nIf you choose to rely on the default value of `shared_buffers` to deduce the shared buffer upper limit, then use the cgroup `memory.max` value to influence the computation.\n\nColumnar engine memory\n----------------------\n\nThe dynamic `shared_buffers` is independent of the [columnar engine memory](/alloydb/omni/current/docs/columnar-engine/configure). When the columnar engine is enabled, the dynamic `shared_buffers` size can be derived by subtracting the amount of memory used by the columnar engine from 80% of the total memory available to the system or cgroup.\n\nHuge pages\n----------\n\nHuge pages improve database performance. AlloyDB Omni manages huge pages explicitly if possible, otherwise it relies on the transparent huge pages (THP) feature of the operating system. If neither huge page type is supported, AlloyDB Omni falls back to 4k page, and prints a warning in docker container logs `docker logs $container_name` with specific instructions to set up huge pages. See [Start AlloyDB Omni](/alloydb/omni/current/docs/run-connect#start) for instructions how to start a container.\n\nThe warning looks similar to the following: \n\n HINT: Please either execute the all-in-one setup script:\n docker run --rm --privileged $image setup-host\n OR manually execute:\n echo within_size | sudo tee /sys/kernel/mm/transparent_hugepage/shmem_enabled\n sudo sysctl -w vm.nr_overcommit_hugepages=1048576\n\nAutomatic memory management at runtime\n--------------------------------------\n\nAlloyDB Omni constantly monitors the system load, and adjusts its memory consumptions for better performance. Specifically, you might observe the following:\n\n**Dynamic `shared_buffers` size change**\n : AlloyDB Omni increases the dynamic `shared_buffers` size when the system memory consumption is low, and decreases the size when the system memory consumption is high. To monitor the dynamic `shared_buffers` size, use: \n\n ```bash\n CREATE EXTENSION IF NOT EXISTS g_memory;\n SELECT g_dynamic_shared_size();\n ```\n\n**Termination of a PostgreSQL connection when the system is extremely low on memory**\n : When AlloyDB Omni detects that the system is extremely low on memory, it attempts to delete the most memory-consuming PostgreSQL connections until the load falls back to a reasonable level. When such an event happens, AlloyDB Omni logs the following in docker container logs: \n\n ```bash\n WARNING: Sending SIGTERM to pid=xxx NSpid=xxx (VA size = xxxMB) (RSS size = xxxMB)\n ```"]]
