The Tool Desk
Outbyte PC Repair FREEClear out junk files and repair common Windows errorsFree Scan →Outbyte Driver Updater FREEScan for outdated or missing drivers - takes under a minuteDriver Scan →High mysqld RAM use is not automatically a leak. MySQL deliberately keeps table and index pages in the InnoDB buffer pool, reserves global caches, allocates buffers for each active connection and statement, and may retain Performance Schema allocations while the server runs. The safe fix is to measure the whole VPS first, identify which category is consuming memory, then make one staged change and verify its effect.
Antihost offers Linux VPS plans with full root access, unlimited traffic on a 1 Gbps port and a 14-day free trial.
Do not begin by blindly lowering every cache or setting an arbitrarily small buffer pool. A VPS also needs memory for the kernel, filesystem cache, web server, application runtime, queues, backups, and monitoring. A setting that looks reasonable for an idle database can still trigger swapping or an out-of-memory kill during a traffic spike.
As an Amazon Associate I earn from qualifying purchases.
What “high memory usage” actually means
Separate three observations before changing configuration:
- Resident process memory: the RAM currently mapped into the
mysqldprocess. - MySQL’s intentional allocations: buffer-pool pages, global buffers, per-session work areas, temporary-table memory, metadata, and instrumentation.
- Host pressure: whether the VPS has little available RAM, is actively using swap, or has killed a process through the Linux OOM (out-of-memory) mechanism.
A large resident set with comfortable available memory and no swap activity may be normal. The same resident set on a VPS that also runs PHP-FPM, a web server, Redis, backup jobs, or monitoring agents can be dangerous. Always compare the database with the host total, available memory, swap behavior, and other services.
#1 Best Overall
- BLUE LIGHT BLOCKING GLASSES: Our blue light glasses for women boast UV400 protection, blocking harmful blue rays from computer screens. With these bluelight glasses women, you can enjoy gaming, reading, or watching TV with no worries of eye strain.
- EYE RELIEF: Our blue light glasses are engineered to reduce blurry vision or strain caused by prolonged screen time. These womens blue light glasses aid in better sleep by preventing blue light from disrupting your natural sleep cycle, helping you fall asleep faster and enjoy a more restful night's sleep, making them the perfect christmas present.
- COMFORTABLE: Crafted from durable material, our trendy blue light glasses for woman or blue light glasses men ensure long-lasting comfort. The high-transmittance lenses accessories, gamer gifts for men, help keep your eyes relaxed. Perfect stocking stuffers, gifts for gamers men, christmas gifts for men, halloween costumes.
- LIGHTWEIGHT: These clear blue light glasses feature a lightweight frame, meaning they’ll stay comfy for long-term wear. Perfect as blue blockers glasses for women or mens blue light glasses, these non-prescription fake glasses for men can be worn throughout the day. The blue light blocking glasses women feature polycarbonate HD lenses, which are anti-reflective and restore true color.
- WORRY-FREE PURCHASE: If you experience any issues, our support team is ready to assist you quickly. These are the perfect costume glasses for cosplay dressing up.
Step 1: Confirm that the VPS is under pressure
Record RAM, swap, and the mysqld resident set
Run these commands during a representative busy period and again when traffic is quiet:
free -h
ps -o pid,ppid,%mem,rss,vsz,cmd -C mysqld
# On systemd hosts:
systemctl status mysql
systemctl status mysqld
# If the service is in a cgroup, inspect its limits and usage:
systemctl show mysql --property=MemoryCurrent,MemoryMax
systemctl show mysqld --property=MemoryCurrent,MemoryMaxfree -h reports total, used, available, and swap figures. “Available” is more useful than the simple used percentage because Linux can reclaim filesystem cache. A rising swap-in or swap-out rate, rather than merely having a swap file present, indicates that RAM pressure is affecting performance.
Look for OOM-killer events
Check the kernel and service logs for evidence that Linux terminated MySQL or another process:
dmesg -T | grep -i -E 'out of memory|oom|killed process'
journalctl -k --since "24 hours ago" | grep -i -E 'out of memory|oom|killed process'
journalctl -u mysql --since "24 hours ago"
journalctl -u mysqld --since "24 hours ago"The exact service name differs by distribution. An OOM event naming PHP-FPM, a backup utility, or another worker is still relevant: that process competed with MySQL for the same VPS memory.
Inventory every resident service
Use ps, your VPS provider’s metrics, or cgroup statistics to list Apache or another web server, PHP-FPM, application workers, Redis, search services, backup compression, log processing, and monitoring agents. Record scheduled jobs that overlap with the incident. MySQL may be innocent of a host-wide shortage caused by a simultaneous backup or deployment.
Step 2: Capture MySQL’s configuration and concurrency
Inspect the main memory-related variables
Save the output before changing anything. Variable names and whether a setting is dynamic vary by MySQL release and distribution.
Rank #2
- WINDEX GAMING WIPES: effortlessly clean all your gaming gear and gaming electronics in one easy step with our new Windex Gaming Wipes
- AMMONIA FREE: our ammonia-free formula is safe to use on all of your favorite gaming equipment like consoles, headsets, controllers, screens, keyboards, and more
- PRE-MOISTENED WIPES: each wipe instantly removes dust, dirt, and stubborn smudges and fingerprints leaving a streak-free shine
- ANTI-STATIC: great for cleaning and wiping down your gaming gear and electronics leaving behind no visible residue
- COMPACT RESEALABLE PACKAGE: store your pack of Windex Gaming Wipes anywhere in your home for whenever a quick and easy wipedown is needed
SHOW VARIABLES WHERE Variable_name IN (
'innodb_buffer_pool_size',
'innodb_log_buffer_size',
'max_connections',
'max_allowed_packet',
'net_buffer_length',
'sort_buffer_size',
'join_buffer_size',
'read_buffer_size',
'read_rnd_buffer_size',
'tmp_table_size',
'max_heap_table_size',
'performance_schema'
);Capture the values with their units and note whether they are global or session-scoped. A global limit is not the same as memory already allocated: many session buffers grow only when a query needs them.
Measure actual connection demand
SHOW GLOBAL STATUS WHERE Variable_name IN (
'Threads_connected',
'Threads_running',
'Max_used_connections'
);Threads_connected is the current number of sessions, Threads_running is the number actively executing, and Max_used_connections records the highest simultaneous connection count since the last server start or status reset. Compare these with max_connections. A very high limit can reserve capacity for a surge that never occurs while allowing a real surge to multiply per-connection memory until the host runs out of RAM.
Estimate concurrency risk without treating limits as a bill
Do not multiply every configured buffer by max_connections and assume that is the exact footprint. Some buffers are allocated lazily, some sessions never use every work area, and allocator overhead is not represented by the variable values. The multiplication is nevertheless a useful worst-case warning: a modest per-session setting can become significant when hundreds of sessions run sort, join, read, or temporary-table operations at once.
Check application pool settings before reducing max_connections. Web workers, job queues, replicas, administrative clients, and health checks all consume connections. Lower the limit only after confirming the application can queue or reuse connections and that its peak demand fits below the new ceiling. Apply the change in a maintenance window if the installed release does not support changing it dynamically.
Quick wins for a faster PC:
Clear out junk files and repair common Windows errorsFree Scan →Fix the driver behind crashes, sound loss and screen glitchesFind Drivers →Repair Windows errors before they cause bigger problemsFix Now →Step 3: Attribute memory with Performance Schema
Check whether memory instruments are available
Performance Schema can allocate memory at startup and while observing workload. It does not return those allocations during normal operation; they are released when the server shuts down. That behavior can make resident memory appear “sticky” even after a workload burst.
SELECT NAME, ENABLED, TIMED
FROM performance_schema.setup_instruments
WHERE NAME LIKE 'memory/%'
ORDER BY NAME;If the table is unavailable, Performance Schema may be disabled or the distribution may expose a different feature set. Confirm the installed version’s manual before enabling instruments or changing startup options.
Read global and per-session summaries
SELECT EVENT_NAME, CURRENT_COUNT_USED, CURRENT_NUMBER_OF_BYTES_USED,
HIGH_COUNT_USED, HIGH_NUMBER_OF_BYTES_USED
FROM performance_schema.memory_summary_global_by_event_name
WHERE CURRENT_NUMBER_OF_BYTES_USED > 0
ORDER BY CURRENT_NUMBER_OF_BYTES_USED DESC;
SELECT USER, HOST, EVENT_NAME, CURRENT_NUMBER_OF_BYTES_USED,
HIGH_NUMBER_OF_BYTES_USED
FROM performance_schema.memory_summary_by_account_by_event_name
WHERE CURRENT_NUMBER_OF_BYTES_USED > 0
ORDER BY CURRENT_NUMBER_OF_BYTES_USED DESC
LIMIT 100;Use the thread summary tables as well when one connection or worker appears responsible. Names such as memory/innodb and memory/sql identify broad code areas rather than a single query. The summaries show instrumented allocations, not the entire process.
Rank #3
- Ergonomic & Adjustable Design: Fight "text neck" and improve your posture. Designed for comfort, it offers an adjustable height (7.1in-8.5in) (1.4in adjustable range) and angle (5°-85°) Please note: cell phone stand is only 9 inches tall—the length of a standard pencil
- Without blocking subtitles Perfect for office desk Tiktok Face Time Zoom video—this is the ultimate kitchen desk essential for phones! Perfect viewing experience while reducing neck back strain, A must-have for gamers streamers desk, and anyone who values both style and support!
- Sleek & Stable Design fits Otterbox Case: Say goodbye to bent cables, tangles, and tip-overs! LISEN’s stylish adjustable desk phone holder features a built-in charging hole and smart cable management, keeping your phone secure and your workspace clutter-free cool gadgets for men
- ANTI-SLIP DESIGN free hands: The under 10.00 items easy gadgets back and the bottom of this kitchen essentials for women are fully covered by anti-skid silicone, which LISEN stand up desk decor can provide maximum protection device from any scratches and slides
- Your Perfect Viewing Companion:If your product arrives damaged or breaks within 365 days, reach out to LISEN for a quick and hassle-free solution. Struggling with assembly, like tightening the screw? We’ve got your back! Your experience matters to us. (Product including: phone Holder stand*1, screw*1, hexagon screwdriver*1, specification*1)
Reconcile instrumentation with the operating-system view
Compare the instrumented total with the resident set reported by ps or cgroup metrics. The gap includes allocator overhead, shared libraries, thread stacks, memory that is not instrumented, and other process areas. A large gap is a reason to investigate further, not proof of a leak. If memory continues to rise and no instrument explains it, treat that as a leak or workload-investigation signal and preserve time-series measurements before restarting.
Step 4: Check whether InnoDB’s buffer pool is the dominant consumer
Understand what the buffer pool does
InnoDB’s buffer pool caches table and index pages so queries can avoid disk reads. Oracle’s MySQL Reference Manual states: “Typically, a recommended innodb_buffer_pool_size value is 50 to 75 percent of system memory.” That is a starting recommendation for a database-focused host, not a promise that the same percentage is safe on a small VPS running an application stack. The pool must leave room for the operating system, connection and statement buffers, temporary tables, logs, and every other service.
Inspect buffer-pool state
SHOW ENGINE INNODB STATUSG
SELECT VARIABLE_NAME, VARIABLE_VALUE
FROM performance_schema.global_status
WHERE VARIABLE_NAME LIKE 'Innodb_buffer_pool%';SHOW ENGINE INNODB STATUSG exposes buffer-pool and activity details in a release-specific format. Use it alongside the configured pool size and host metrics. A pool consuming most of the VPS while application services share the machine is a likely pressure source; a small pool with a large resident set points elsewhere.
Reduce the pool in a controlled step
First verify that the installed MySQL release supports dynamic resizing and confirm the exact syntax in that release’s documentation. If it does, reduce the value by one measured step rather than cutting it to an arbitrary fraction. Observe available memory, swap, read latency, query latency, and error rates through a representative workload. If the setting requires a restart, schedule one, record the old value, and make sure the configuration file and startup automation agree so the change is not accidentally reverted.
A smaller pool protects the whole VPS but can increase disk reads and page churn. A pool that is too large can cause swapping; one that is too small can cause excessive churn. The correct value is the largest one that leaves dependable headroom for the host and workload, not the largest value the database can technically allocate.
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
Step 5: Control connection and statement buffers
Why a single session can grow beyond its initial allocation
Connection and result buffers start at net_buffer_length and can grow up to max_allowed_packet. Each session also receives max_digest_length bytes for statement digests. Sort, join, read, and read-rnd buffers are additional per-session work areas. These allocations can multiply across concurrent sessions, especially when many complex queries run simultaneously.
Rank #4
- Easy Installation: Our gaming lights have been designed to enhance your entertainment experience. Simply position the led light bar next to your TV or game console to create a game room environment with ease. They come with USB power and a 60 inch cord for direct mounting and mobility
- Audio Syncing & Extensive Mode: The built-in high-sensitivity music sensor allows you to sync the TV led lights to your favorite music. Lights react and move to the beat of music, film and game audio, creating an incredible immersive sound and light experience – and all without the need for an additional hub. You can choose from 64+ patterns, including strobe, illusion, jump, energy, gradient, running, chasing, fluttering mode etc. Each mode can be applied to different scenes
- 64+ Chasing IC+RGB & W Effect: The independent control design of IC+RGB and white all-in-one multifunctional rgbw led bar lights monitor backlight allows you to easily switch the led lights for desk between IC+RGB and white to enjoy a variety of lighting effects, you can choose white lighting mode with high brightness for better illumination when working, etc. Choose IC+RGB lighting mode when you need to create an entertaining atmosphere for the light show, 16 million colours of Color changing rgb light bar let you dance to the beat and create a dynamic party atmosphere
- Dynamic and Dimmable Lights: 1%-full brightness free brightness adjustment gives you the flexibility to adjust dimmable led light bar for room decorations or a variety of occasions
- USB Powered: USB light bars for room can be powered by powernank, usb charger, TV, Monitor, laptop PC mainframe, making behind TV led lighting bar easy to install in any location.TV led lights are powered by the TV usb interface / PC mainframe, can be turned on/off synchronously with TV / PC
Inspect the variables rather than assuming distribution defaults:
SHOW VARIABLES WHERE Variable_name IN (
'net_buffer_length',
'max_allowed_packet',
'max_digest_length',
'sort_buffer_size',
'join_buffer_size',
'read_buffer_size',
'read_rnd_buffer_size'
);Use workload-driven limits
Keep max_allowed_packet large enough for the largest legitimate row or statement sent by the application, but do not use a very large value as a substitute for fixing oversized requests. Set sort and join buffers for the queries that actually need them; increasing them globally can make a connection spike much worse. Prefer query plans and indexes that avoid large in-memory sorts or joins. If only one reporting job needs a larger session value, set it for that job’s session instead of raising the global default.
After changing a per-session or global value, watch both peak resident memory and query behavior. A lower buffer may reduce RAM while causing disk-based operations or slower queries. That trade-off is acceptable only if the VPS remains stable and the resulting latency meets the application’s requirement.
Step 6: Prevent internal temporary tables from exhausting RAM
Identify temporary-table pressure
Queries that build internal temporary tables can consume memory up to the applicable limits. Google Cloud’s MySQL guidance recommends limiting tmp_table_size when internal temporary tables use too much memory. Check both limits because an internal in-memory table is constrained by the lower effective setting in the relevant execution path:
SHOW VARIABLES WHERE Variable_name IN (
'tmp_table_size',
'max_heap_table_size'
);Correlate memory summaries with temporary-table status counters and slow-query data. Look for reports, sorts, grouping, distinct operations, and joins that create large intermediate results. Fixing the query or adding an appropriate index is usually more durable than merely lowering a limit.
Lower limits with the I/O trade-off understood
Reduce tmp_table_size and max_heap_table_size in a staged change when temporary tables are a confirmed source of spikes. Oversized results will spill to disk sooner, which protects RAM but can increase disk I/O and latency. Make sure the VPS has sufficient disk space and acceptable storage performance, then measure query time and I/O wait after the change. Do not claim success solely because the resident set fell; a database that is swapping less but spending all its time in disk I/O still needs query work.
Free tools Windows power users keep installed
One-click scans. No signup required.
Best Value
- Compatible with Wide Screens - To ensure compatibility with the dual monitor mount, your each monitor must meet three conditions at the same time: First, computer screens size range: 13 to 32 inches. Second, screen weight range: 4.4 to 19.8 lbs. Third, the back of the monitor screen must have VESA mounting holes with a pitch of 75x75mm or 100x100mm.
- Regarding the compatibility with desks - Your desk must meet three conditions at the same time: First, desk material: Only wooden desks are recommended, plastic or glass desks cannot be used. Second, desk thickness range: 0.59" - 3.54". Third, the bottom of the desk should not have any cross beams or panels, as this will interfere with installation. We recommend carefully checking that your desk and monitors meets all above conditions before purchasing.
- Dual C-Clamp Hold - Worried your dual monitors might wobble or slip? Our upgraded base uses a larger platform plus a dual C-clamp structure to lock the dual monitor arm firmly to your desk. Each arm safely keeps your screens steady while you type, click and game—no shaking, no sliding, just a clean and secure setup you can trust every day. It also provides Grommet Mounting installation choice, both options ensure stable and secure fixation for your 0.59" - 3.54" desk.
- Full-Motion Adjustment For Comfortable View - Pull the screen closer when you’re deep in a spreadsheet, push it back to watch videos, or rotate to portrait for coding — moving everything smoothly with just one hand. The monitor stand offers +85°/-50° tilt, ±90° swivel and 360° rotation. Raise your monitor up to 15.75″ to support a healthy sitting posture. Whether you’re working from home, gaming through the night, or switching between video calls and documents, getting the screens to your natural line of sight helps relieve neck, shoulder and back strain so you can stay focused longer with less fatigue.
- Keep Your Desk Organized: By lifting both screens off the desktop, this dual monitor stand opens up valuable space for your keyboard, notebook, docking station or a simple, clutter-free work area. Built-in cable management guides wires along the arms, keeping cords out of sight and out of the way. Enjoy a tidy, modern workstation that looks as good as it feels to use.
Step 7: Investigate Performance Schema growth without disabling useful observability
Performance Schema may allocate autoscaled buffers as it encounters new accounts, threads, digests, or instruments. It retains those allocations until shutdown. Do not disable instrumentation blindly: first use the memory summaries to identify the category. Where the installed release provides sizing parameters for autoscaled buffers, cap only the categories that are growing beyond their diagnostic value and confirm the release-specific variable names.
A restart releases Performance Schema memory, so a controlled restart can contain a one-time accumulation after you have captured diagnostics. It is not a permanent fix. If the same category grows again under the same workload, investigate connection churn, statement-digest cardinality, account or thread counts, and the workload pattern that creates those objects.
Step 8: Validate every change under representative load
- Record a baseline. Save
free -h, themysqldresident set, swap activity, OOM logs, MySQL variables, connection status, query latency, error rates, and disk I/O. - Change one category. For example, adjust the buffer pool or temporary-table limits, but not all memory variables at once.
- Use a controlled window. Observe a normal traffic period and the known peak, including scheduled jobs if they contributed to the incident.
- Check both safety and performance. Confirm that available memory rises, swap-in/out falls or stays quiet, no OOM event appears, and latency and errors remain acceptable.
- Keep a rollback value. Revert the last change if query latency, disk I/O, or application errors worsen beyond the agreed threshold.
- Document the reason. Record the observed consumer, old and new value, date, release, and measured result so a package upgrade or configuration rebuild does not erase the decision.
Choosing among the main fixes
| Fix | Peak-memory effect | Likely trade-off | Restart requirement | Protects |
|---|---|---|---|---|
Reduce innodb_buffer_pool_size |
Reduces a large global allocation | More disk reads and possible page churn | Release-dependent; verify dynamic resizing | The whole VPS when the pool crowds out other services |
Lower max_connections |
Limits the concurrency multiplier | Clients may be refused or queued if application pools are mis-sized | Release-dependent | The host from connection storms |
| Trim per-session buffers | Reduces possible memory per active query | Sorts and joins may become slower or spill to disk | Often session or global behavior varies by variable | The host during concurrent work |
| Lower temporary-table limits | Caps in-memory intermediate results | More temporary disk I/O and latency | Release-dependent | RAM during large reports and aggregations |
| Cap Performance Schema autoscaling | Stops selected instrumentation categories growing without bound | Less history or detail for those categories | Some settings require restart | MySQL’s instrumentation footprint |
| Add VPS RAM or move MySQL | Increases available headroom rather than shrinking MySQL | Higher recurring cost and possible migration work | Provider- or migration-dependent | The complete service stack |
Use the table only after attribution. Resizing a VPS is sensible when measured, safe settings still leave no headroom; it is not a substitute for an uncontrolled connection storm or a query that creates enormous temporary results.
A practical sizing approach for a small VPS
Start with the VPS’s actual usable memory, then reserve a fixed margin for the operating system and every non-MySQL service. From what remains, account for the InnoDB pool, global log and key buffers, peak connection work areas, temporary tables, Performance Schema, and allocator overhead. Recalculate this budget when you add workers, replicas, backup compression, or monitoring.
Do these 3 things before closing this tab:
1Repair Windows errors before they cause bigger problems2Fix the driver behind crashes, sound loss and screen glitches3Clear out junk files and repair common Windows errorsThe documented 50–75 percent buffer-pool guidance can be a useful upper starting range on a database-focused machine, but a gaming panel, web application, or control-plane services on the same VPS may require a materially smaller share. Use observed peak concurrency and available-memory measurements to choose the value. Never size from idle memory alone.
When a restart is justified—and when it is not
Restarting MySQL releases memory held by Performance Schema and can clear fragmentation or a one-time accumulation, but it interrupts service and hides evidence if performed before measurement. Capture OS metrics, status variables, Performance Schema summaries, logs, and the current configuration first. Restart as controlled containment when memory growth threatens the host or when a documented setting requires it; then monitor whether the same allocation pattern returns.
If growth returns after attribution and staged tuning, investigate the workload, connection churn, statement patterns, and possible defects in the installed release. A persistent unexplained increase that survives normal workload changes warrants escalation to the distribution or MySQL support channel with timestamps, configuration, status output, and memory profiles.
When to increase capacity or separate MySQL
Increase VPS RAM or move MySQL to a managed or dedicated database when the measured minimum headroom remains too small after reasonable tuning, or when the application’s legitimate peak cannot fit alongside the operating system and required services. Capacity is the correct fix when demand is real; it is the wrong first response to a mis-sized max_connections, runaway temporary-table query, or unexplained instrumentation growth.
Recommended Free Tools
Before migrating, record peak connections, buffer-pool needs, disk I/O, backup windows, and replication or durability requirements. Those measurements determine the target size and expose whether the current problem is RAM, storage latency, or both.
Quick Recap
Emergency recovery checklist
- Confirm whether the host is swapping or logging OOM kills.
- Stop or reschedule nonessential backup, batch, or monitoring jobs that are competing for RAM.
- Use the application’s connection pool controls to stop a connection storm; do not repeatedly restart MySQL without addressing the source.
- Capture the current process, host, MySQL status, and Performance Schema evidence if the server is still responsive.
- If the service is at immediate risk, apply the one previously tested emergency limit or perform a planned restart, then verify recovery.
- After stability returns, perform the full attribution sequence and set a documented rollback plan.
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
