Run uptime for a quick reading, top for live task and CPU details, or cat /proc/loadavg for the kernel’s raw values. Linux reports three load averages—covering the last 1, 5, and 15 minutes. They count work that is runnable or waiting in uninterruptible I/O, not CPU utilization percentages. To judge whether a value is concerning, compare it with the server’s logical CPU count, inspect the trend, and then identify whether CPU, storage, memory, or another bottleneck is creating the queue.
Antihost offers Linux VPS plans with full root access, unlimited traffic on a 1 Gbps port and a 14-day free trial.
What load average means
Linux load average is an exponentially smoothed estimate of how many scheduling entities are demanding service over time. The kernel includes tasks in the runnable state (R) and tasks in uninterruptible sleep (D), typically waiting for disk or another I/O operation that cannot be interrupted safely.
As an Amazon Associate I earn from qualifying purchases.
A runnable task is either executing on a CPU or ready to execute but waiting for one. A task in D state may consume little or no CPU at the instant you look at it, yet it still contributes to load because it is waiting for access to a resource. That is why a server can show a high load average while CPU utilization appears modest.
The three values are time windows, not percentages:
#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.
| Value | Window | What it helps show |
|---|---|---|
| First | 1 minute | Recent pressure or a short-lived spike |
| Second | 5 minutes | Short-term persistence |
| Third | 15 minutes | Longer-running pressure or recovery from earlier work |
The Linux proc_loadavg(5) documentation defines these fields as the number of jobs in the run queue (state R) or waiting for disk I/O (state D), averaged over 1, 5, and 15 minutes. The uptime(1) documentation describes the same population as processes in runnable or uninterruptible state.
Commands that show the load average
Use uptime for a quick check
uptimeA typical result looks like this:
14:32:10 up 12 days, 3:18, 2 users, load average: 1.20, 0.85, 0.60The three numbers at the end are the 1-, 5-, and 15-minute averages, in that order. The command also gives the current time, how long the system has been running, and the number of logged-in users; those other fields do not affect the load calculation.
Do these 3 things before closing this tab:
1Clear out junk files and repair common Windows errors2Fix the driver behind crashes, sound loss and screen glitches3Repair Windows errors before they cause bigger problemsUse w when you also need logged-in sessions
ww includes the same load-average summary as uptime, followed by users and their current processes. It is useful during an incident when you want to see whether an interactive session, shell, or administrator is active on the host.
Use top for a live view
topThe summary area at the top of top shows the 1-, 5-, and 15-minute values. Below it, the task list lets you correlate load with process state and resource consumption. Press q to leave.
Inside top, sort by CPU usage when looking for active consumers, but do not assume the highest CPU process is responsible for all load. Processes in D state can raise load while using little CPU at that moment.
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
Read the kernel interface directly
cat /proc/loadavgA line may look like:
1.20 0.85 0.60 3/842 27194The fields are:
| Field | Meaning |
|---|---|
| 1 | 1-minute load average |
| 2 | 5-minute load average |
| 3 | 15-minute load average |
| 4 | Currently runnable scheduling entities divided by total scheduling entities |
| 5 | Process ID (PID) of the most recently created process |
The fourth field is a snapshot, unlike the smoothed averages. The numerator counts entities currently runnable; the denominator is the total number of scheduling entities. The final PID is informational and is not a measure of load.
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
How to judge a number on a particular server
Find the logical CPU count
nprocFor a fuller CPU description, use:
lscpu | grep '^CPU(s):'These commands report logical CPUs—the execution units visible to Linux, including hardware threads where simultaneous multithreading is enabled. Compare load with that number, not automatically with the number of physical cores.
Normalize load by logical CPUs
A load of 4 means something different on a one-CPU host than on a 16-CPU host. A practical ratio is:
load average ÷ logical CPU countFor example, a 2.0 one-minute load is approximately 200% of the runnable capacity of a one-logical-CPU host, but only 25% of the logical-CPU count on an eight-logical-CPU host. This ratio is a diagnostic aid, not a universal pass/fail rule.
| Host | 1-minute load | Load ÷ logical CPUs | Initial interpretation |
|---|---|---|---|
| 1 logical CPU | 2.0 | 2.0 | Work is likely queuing if the condition persists |
| 4 logical CPUs | 4.0 | 1.0 | Runnable capacity is approximately fully occupied |
| 8 logical CPUs | 4.0 | 0.5 | Load is below the logical-CPU count; investigate impact and I/O state before calling it a problem |
Values near the logical-CPU count indicate that runnable capacity is being fully used. Sustained values above that count indicate that work is queuing somewhere. Linux’s manuals define what gets counted but do not establish one number that is always “too high.” Workload, latency requirements, task mix, and storage behavior determine whether a given value is harmful.
Outdated Drivers Are Slowing You Down
One free scan finds every outdated or missing driver and matches the right update for your exact hardware.Free scan · exact hardware matchWindows Errors? Fix Them Before They Spread
Repair common Windows errors and clear accumulated junk for a smoother, more stable PC - no reinstall needed.Free scan · no reinstallRank #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)
Read the 1-, 5-, and 15-minute trend
Always read the numbers in order: 1 minute, then 5 minutes, then 15 minutes.
- 1 minute much higher than 5 and 15 minutes: pressure increased recently. This often points to a burst, a newly started job, a traffic surge, or a short I/O stall. Check again and correlate it with the event rather than reacting to one sample.
- 1 minute lower than 15 minutes: the host was busier earlier and is now recovering. A falling 1-minute value does not prove that queues and user-visible latency have cleared.
- All three elevated and close together: pressure has been sustained for at least several minutes. Inspect active tasks, resource counters, and application impact promptly.
- All three low: there is little qualifying runnable or uninterruptible work averaged over those windows. This does not rule out a brief spike, a latency problem that does not create many queued tasks, or a failing application.
These are interpretations of the documented windows, not extra kernel metrics. Confirm them with task states and CPU, memory, and storage data.
A repeatable diagnostic sequence
1. Capture load and capacity together
date; uptime; nproc; cat /proc/loadavgRecording the timestamp, load windows, CPU capacity, and raw kernel line in one snapshot prevents confusion when comparing hosts or reviewing an incident later.
2. Inspect runnable and uninterruptible tasks
topLook at the task-state column and identify processes or threads marked R and D. Sort by CPU to find active consumers, but remember that a D task may be blocked in I/O rather than running. A long list of R tasks suggests competition for CPU time; many D tasks suggest a blocked I/O path or another kernel operation that is waiting without interruption.
Free tools Windows power users keep installed
One-click scans. No signup required.
3. Separate CPU pressure from I/O pressure
In top, examine the CPU summary fields:
- user: time spent running user-space programs.
- system: time spent in the kernel on behalf of programs.
- idle: time when CPUs had no runnable work.
- iowait: time CPUs were idle while waiting for I/O completion, where the platform reports it.
- steal: time taken from a virtual machine by the hypervisor, where the platform reports it.
High load with little idle time and substantial runnable work points toward CPU contention. High load accompanied by many D tasks or elevated I/O-wait points toward storage or another uninterruptible I/O path. High steal time suggests that the virtual machine is waiting for host-level CPU scheduling rather than simply running too much application code.
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
Linux user-space tools derive CPU-state measurements from counters such as /proc/stat and /proc/uptime. Those counters describe CPU time states; they do not replace the load average, because load also includes uninterruptible tasks.
4. Check memory pressure
free -h
vmstat 1free -h gives a human-readable memory summary. vmstat 1 prints successive one-second samples; watch for sustained swapping or paging activity, blocked processes, and unusual memory movement. Memory exhaustion can trigger reclaim and I/O that leave tasks waiting in D state, so a storage-looking load problem may begin with insufficient memory.
5. Check storage and device queues
iostatiostat is supplied by common system-statistics packages and may not be installed by default. If available, use its device and utilization data to determine whether a disk or virtual block device is saturated, experiencing high latency, or building a queue. Match the device with the processes shown in top; a high load average alone cannot identify the responsible device or program.
6. Correlate with real impact and duration
Compare the load trend with request latency, queue depth, error rates, login responsiveness, deployment times, and scheduled batch jobs. A short spike with no user-visible effect may require observation rather than emergency action. Sustained saturation combined with rising latency or errors requires remediation and capacity planning.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.Why high load can coexist with low CPU usage
CPU percentage and load average answer different questions. CPU utilization asks how much time processors spent in particular states. Load average asks how many tasks were runnable or uninterruptibly waiting, averaged over time.
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.
Consider a process blocked on storage in D state. It contributes to load while it waits, but it is not consuming CPU during that wait. Several blocked readers or writers can therefore push load upward while top shows significant idle time. Similarly, a virtual machine may report high steal time because the hypervisor is withholding CPU; the guest’s workload and its observed CPU counters then need to be interpreted together.
Do not “fix” this pattern by killing the process with the highest CPU percentage unless task and I/O evidence identifies it as the cause. The highest-CPU process may be unrelated to the blocked work that is inflating load.
Common interpretations and examples
Example: a recent CPU burst
A four-logical-CPU server reports 8.0, 2.1, 1.0. The 1-minute value is far above the older windows, so pressure rose recently. If top shows many R tasks, little idle time, and one or more CPU-heavy jobs, investigate the burst source—such as a batch process or traffic increase—and watch whether the value falls as the work completes.
Example: sustained storage waiting
An eight-logical-CPU server reports 6.4, 6.1, 5.9, while CPU idle remains substantial and many tasks are in D. The load is below the CPU count, but the pattern is persistent and the task state points to an I/O bottleneck. Check iostat, device latency, queue depth, filesystem health, and any backup, scan, or batch activity that coincides with the wait.
Example: recovery after an incident
A host reports 1.5, 4.2, 7.8. The one-minute value is well below the 15-minute value, indicating that the system was busier earlier. Confirm that queues, latency, and errors are also declining; the falling average by itself does not prove that every request has recovered.
What load average cannot tell you
- It does not identify a culprit process. The value aggregates qualifying scheduling entities over time.
- It does not state which device, filesystem, network path, or external service is slow.
- It does not provide CPU utilization as a percentage.
- It does not account for your application’s acceptable latency or throughput target.
- It does not reveal a short event precisely; the 1-minute figure is still averaged over a window.
Use it as an early signal that work is waiting or competing, then use task states and resource-specific counters to find the constraint.
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 →Frequent mistakes to avoid
- Treating load as a percentage: a value of 1.0 is not “1% CPU.” It is an average count of qualifying work.
- Comparing every host with 1.0: always record logical CPU count and normalize the value.
- Ignoring
Dstate: low CPU utilization does not rule out storage or uninterruptible-wait pressure. - Reacting to one sample: compare all three windows and take another snapshot before deciding whether a spike is transient.
- Assuming a threshold is universal: a latency-sensitive service may suffer at a lower ratio than a batch host, while a compute-heavy host may tolerate brief higher values.
- Reading only the load line: inspect CPU fields, memory, storage, task states, and application symptoms before choosing a remedy.
A compact command checklist
- Run
uptimeto see the three averages immediately. - Run
nprocand compare each average with the logical-CPU count. - Run
cat /proc/loadavgwhen you need the raw fields, including runnable/total entities and the newest PID. - Open
top; inspectRandDtasks and the CPU summary. - Run
free -handvmstat 1for memory and paging context. - Use
iostatwhen storage latency or queueing is suspected and the utility is installed. - Correlate the measurements with latency, errors, queue depth, and the timing of jobs or deployments.
When to take action
Take action when elevated load is sustained, exceeds the host’s available capacity for the workload, or coincides with user-visible latency, errors, or growing queues. The remedy depends on the evidence: reduce or reschedule CPU-heavy work when runnable contention is confirmed; investigate disks, filesystems, and I/O paths when D tasks dominate; address memory pressure when swapping or reclaim is driving waits; and investigate host contention when virtualization steal time is high.
A brief, isolated spike without impact can be logged and monitored. The important decision is not whether a number looks large in isolation, but whether the trend represents a constrained resource and whether that constraint is affecting the service.
Quick Recap
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.
