Hosting resource limits can look technical, but they are practical signals about how much work your website can handle. This guide explains CPU, RAM, storage, inodes, bandwidth, PHP workers, database limits, and throttling messages, then gives you a simple review process for deciding whether to optimize, upgrade, or migrate.
Overview
Every web hosting plan places boundaries on shared resources. The exact limits and labels vary by provider, so a number shown in one hosting dashboard may not be directly comparable with a number from another. Treat the figures as operating constraints, not as a universal hosting scorecard.
A resource warning does not automatically mean that your website needs a more expensive plan. A badly configured plugin, an unoptimized image library, a traffic spike, or an inefficient database query can create a short-term problem. On the other hand, repeated warnings during normal traffic may indicate that your current hosting plan no longer matches the site.
The useful question is not simply, “How large is the limit?” It is, “Which resource is being exhausted, when does it happen, and what is causing the demand?” Record those details before making a change.
For a broader plan comparison, see Shared Hosting vs VPS. If a resource issue follows a recent DNS or server change, the DNS troubleshooting guide can help separate hosting performance problems from connection and configuration problems.
What to track
Start by recording the resource name, the provider’s stated limit, your observed usage, the time period, and the effect on visitors. Screenshots or exported reports are useful because dashboard data may reset after a billing period or reporting window.
| Resource | What it measures | Signs of pressure | What to check first |
|---|---|---|---|
| CPU usage or CPU seconds | How much processor time account processes consume. Some hosts show a percentage; others report accumulated CPU seconds over a period. | Slow dynamic pages, processes being limited, or notices that account CPU usage is too high. | Traffic spikes, scheduled tasks, plugins, themes, bots, and slow application code. |
| RAM or physical memory | Memory used while PHP, database, control-panel, or other processes are running. | Failed requests, memory-limit errors, unstable administration screens, or processes being stopped. | PHP memory settings, large imports, plugins, simultaneous requests, and application logs. |
| Storage | Disk space used by website files, media, email, databases, logs, backups, and temporary files. | Failed uploads, backup errors, full-mailbox warnings, or inability to update software. | Old backups, large media files, logs, staging copies, and unused email. |
| Inodes | The number of files, folders, emails, and other filesystem entries in an account. | File-creation failures even when available disk space appears sufficient. | Cache files, image variants, session files, email, backup archives, and abandoned installations. |
| Bandwidth or data transfer | The amount of data sent to and from the website, often including page assets, downloads, media, and email. | Transfer warnings, reduced service, or unexpected usage during a traffic or download event. | Large images, video, downloads, hotlinking, crawlers, and traffic sources. |
| PHP workers | The number of PHP requests that can run at the same time for a site or account. | Requests waiting in a queue, intermittent timeouts, or pages that load only after a delay. | Slow uncached pages, checkout or search activity, external API calls, and traffic concurrency. |
| Database limits | Restrictions may involve database size, connections, queries, tables, or account-level databases. | Database connection errors, slow queries, failed imports, or inability to create or update data. | Database size, autoloaded data, revisions, inefficient queries, and connection-heavy extensions. |
CPU and CPU seconds
CPU measures processing work. A site can use substantial CPU without serving many visitors if a scheduled task repeatedly scans files, a plugin performs expensive calculations, or a crawler requests uncached pages. “CPU seconds” usually describes accumulated processor time during a reporting period, but the measurement window and enforcement method depend on the provider.
Do not compare CPU figures without checking the unit, time window, and account scope. Ask whether the dashboard reports your website alone or all services under the hosting account.
RAM and PHP workers
RAM and PHP workers are related but different. RAM concerns how much memory running processes consume. PHP workers concern how many dynamic requests can be processed concurrently. A page can be memory-heavy, worker-heavy, or both. Increasing workers may not help if each request is slow or consumes too much memory; it can sometimes increase total memory demand.
Storage, inodes, and bandwidth
Storage is measured by size, while inodes count entries. A site with many small cache files can approach an inode limit before using all available gigabytes. Conversely, a media-heavy site may use substantial storage with relatively fewer files. Bandwidth describes transfer, so a single large downloadable file or an embedded video can affect it quickly even when the site itself is small.
Throttling messages
Messages such as “resource limit reached,” “account throttled,” “entry processes exceeded,” or “IO limit reached” are provider-specific. Save the exact wording and note the timestamp. The message may identify the affected resource, but it may not identify the cause. Look for access logs, error logs, cron reports, application monitoring, and traffic analytics before deleting files or changing plans.
Cadence and checkpoints
A monthly review is suitable for a growing site, while a quarterly review may be enough for a stable brochure site. Review more frequently during a launch, migration, campaign, seasonal traffic period, or major WordPress change.
Use this worksheet for each review:
- Record the date and reporting period. Include the hosting plan and the dashboard location where the data was found.
- Capture limits and usage. Write down CPU, RAM, storage, inodes, bandwidth, PHP workers, database figures, and any process or file-count limits that the provider displays.
- Mark peaks, not just averages. A monthly average can hide a short period of overload. Note the busiest day, the time of day, and whether the event was expected.
- Match usage to site events. Compare warnings with publishing, promotions, backups, imports, updates, crawls, or unusual referral traffic.
- Record visitor impact. Note slow pages, failed forms, checkout errors, unavailable administration screens, or reports from users.
- Choose one next action. Optimize, monitor, ask the host for clarification, upgrade temporarily, or plan a migration. Avoid making several untracked changes at once.
Keep the worksheet with your website backup and operational records. The website backup strategy guide covers the separate task of ensuring that a resource problem does not become a recovery problem.
How to interpret changes
Optimize first when the increase is sudden. A new plugin, theme, import, backup job, cache configuration, or bot pattern may explain a sharp change. Review recent changes and test them on a staging copy where possible. Remove unused files carefully, compress oversized media, improve caching, and investigate slow queries rather than treating the limit as the root cause.
Monitor when usage is rising gradually but the site remains stable. Growth is not automatically a problem. Establish a personal warning level below the provider’s hard limit and watch the trend across several reporting periods. A site that regularly approaches a limit has less room for a campaign, traffic burst, or maintenance task.
Upgrade when demand is normal, recurring, and linked to business activity. An upgrade may be reasonable when visitors consistently experience delays, warnings recur after sensible optimization, or the site needs more concurrent PHP processing, memory, storage, or database capacity. Confirm which resource the new plan increases; a larger storage allowance will not necessarily solve a PHP worker or CPU constraint.
Migrate when the constraint is structural. Consider a different hosting environment when the application requires controls, performance characteristics, isolation, or configuration options that the current plan cannot provide. Plan the move around backups, DNS, email, SSL, database compatibility, and rollback. The domain transfer checklist is relevant if the domain registration will move as part of the project, but changing hosts and transferring a domain are separate decisions.
Be cautious with unexplained “unlimited” labels. A plan may still have practical limits on CPU, memory, inodes, processes, database activity, or acceptable usage. Read the resource panel and service terms together, and ask support for the measurement period when the terminology is unclear.
When to revisit
Revisit this review after any resource-limit warning, a noticeable change in page speed, a traffic increase, a WordPress update, a new ecommerce or membership feature, a large media upload, or a hosting renewal. Also review after migration, because the new provider may measure CPU, memory, workers, and storage differently.
At minimum, schedule a monthly or quarterly checkpoint. Compare the current worksheet with the previous one and look for direction rather than a single alarming number. Ask:
- Which resource changed most?
- Was the change temporary or repeated?
- Did users experience a real failure or only an internal warning?
- What site event explains the change?
- Can the cause be fixed without reducing a necessary feature?
- Would an upgrade solve the specific constraint, or is migration more appropriate?
Set a calendar reminder, save the latest dashboard evidence, and keep a short change log. When the answer is unclear, contact the host with the exact warning, timestamps, affected domain, and recent site changes. Clear records turn a vague hosting complaint into a useful performance investigation.
Finally, keep plan comparisons current. Providers can change dashboard terminology, included resources, and enforcement methods. Recheck the figures before renewal or migration rather than relying on an old comparison. For the related operational side of ownership, see the domain security checklist and the website launch checklist.