To add RAM to a hosted VPS, resize its virtual machine plan or replace it with a larger plan from the provider. You do not install a physical memory module, and adding a block-storage volume does not increase RAM. Most upgrades require a shutdown, reboot, migration, or replacement, so take a snapshot or backup first and plan for a short outage.
Need a server to try this on? Antihost offers Linux VPS plans with NVMe storage, full root access and a 14-day free trial.
What “adding RAM” means on a VPS
A VPS receives its memory allocation from the provider’s virtual-machine plan. The hypervisor assigns that amount of RAM to your instance; there is no physical DIMM for you to install. Increasing memory therefore means changing the plan, resizing the instance, or creating a replacement instance from a snapshot with a larger bundle.
As an Amazon Associate I earn from qualifying purchases.
Free tools Windows power users keep installed
One-click scans. No signup required.
The exact operation depends on the provider. Some services offer an in-place resize, sometimes with a RAM-and-CPU-only option. Others use fixed bundles and require a snapshot-based replacement. A larger plan can also change vCPU count, storage capacity, transfer allowance, or the monthly price. Read the plan details before confirming the change.
DigitalOcean describes the benefit directly: “Increasing a Droplet’s memory and CPU improves its performance.” That statement applies when memory pressure is actually limiting the workload; buying RAM will not fix a CPU-bound process, slow storage, a database misconfiguration, or an application leak.
#1 Best Overall
- Capacity: 16GB (2x 8GB Modules) | Type: DDR3 240-Pin | Speed: 1600MHz PC3-12800 / (PC3-12800E) | ECC Type: ECC-UDIMM (ECC Unbuffered DIMM) | Rank: 2Rx8 (Dual Rank x8) | Voltage: 1.35V
- Designed for ECC UDIMM Compatible Servers/Workstations (Rated Speeds & ECC Capabilities are CPU Dependent). Not Compatible with Desktops/Laptops.
- ECC Types can not be mixed | All installed modules must be ECC UDIMMs in order to function properly | A maximum of eight ranks per memory channel can be installed at once
- All A-Tech memory modules undergo stringent quality control testing to ensure dependable and reliable performance
- Backed by A-Tech's Limited Lifetime Warranty + Tech Support Team available to help before and after your purchase
Diagnose memory pressure before you pay for a larger plan
First establish that RAM is the bottleneck. A server can feel slow while still having available memory because of CPU saturation, disk I/O, a database lock, a PHP worker limit, a container limit, or a faulty application.
Check current RAM and swap
- Connect to the VPS over SSH.
- Run
free -h. Note total, used, available, and swap values. The available column is more useful than “free” alone because Linux uses otherwise idle RAM for cache. - Run
vmstat 1 5. Sustained swap-in or swap-out activity, shown in thesiandsocolumns, indicates that the system is moving pages between RAM and disk. - Run
swapon --showto list active swap devices or files. - Review application and kernel logs for out-of-memory events. On many Linux distributions,
journalctl -kor the system log will show an OOM-killer message and the process that was terminated.
Check limits outside the VPS plan
Containers, hosting control panels, systemd services, and application pools can each impose a lower memory ceiling than the virtual machine. A container may be killed while the host still reports available RAM. Check the container’s memory limit, the service unit’s resource settings, and limits for PHP-FPM, Node.js, Java, or other runtimes before changing the VPS plan.
Distinguish RAM pressure from other bottlenecks
- CPU saturation: high load with consistently busy vCPUs points to a compute bottleneck, not necessarily a memory shortage.
- Disk I/O: high I/O wait or a full volume can make every request slow even when RAM is available.
- Database configuration: an oversized buffer pool, too many connections, or inefficient queries can consume memory or CPU.
- Worker limits: web-server or PHP worker counts can cap throughput independently of installed RAM.
- Memory leaks: a process that grows continuously may eventually exhaust any fixed plan; upgrading buys time but does not correct the leak.
A documented example comes from small Amazon Lightsail WordPress instances: when memory is exhausted, the Linux kernel may kill MariaDB. If your logs show that pattern, reducing database and worker usage or moving to a larger bundle is more appropriate than simply adding storage.
Choose the right upgrade size
Measure peak usage during the busiest period, then leave headroom for deployment, backups, traffic spikes, and background jobs. Selecting only the next size can result in another disruptive resize if the workload is growing rapidly. Plan for the memory required by the operating system, database, web server, application workers, monitoring agents, and cache simultaneously.
| Decision | Use it when | Important consequence |
|---|---|---|
| RAM-only or CPU-and-RAM resize | You need memory and perhaps more vCPUs, but existing disk capacity is adequate. | Disk size may remain unchanged, depending on the provider and plan. |
| CPU, RAM, and disk resize | You need more memory and expect storage growth or require a combined plan. | The disk allocation can increase with the plan and may not be reducible later. |
| Snapshot-based replacement | The provider uses fixed bundles rather than in-place resizing. | The new instance can have a different public IP unless you remap a static address. |
| Swap tuning | You need a temporary safety buffer on a small-memory instance. | Swap uses disk and is far slower than RAM; it is not an increase to the plan’s memory. |
Plan prices and available sizes vary by provider, region, operating system image, billing term, and date. Confirm the current plan specification and recurring charge in the provider console before applying an upgrade.
Prepare safely before resizing
- Record the baseline. Save the output of
free -handswapon --show. Record disk usage, mounted volumes, service status, firewall rules, DNS records, public and private IP addresses, and any scheduled jobs. - Take a fresh snapshot or backup. A snapshot gives you a rollback point if the resize or replacement fails. For databases, also keep a current logical or physical backup; do not rely on an old image alone.
- Schedule a low-traffic window. Expect a reboot, migration, or replacement. Tell users about the maintenance period and pause jobs that could be interrupted.
- Check application dependencies. Confirm that the database, web server, queues, certificates, mounts, and monitoring agents will start on the upgraded instance.
- Plan network continuity. Determine whether the public IPv4 address is preserved. If it is not, prepare DNS changes or reserve a static address before the cutover.
- Choose a plan with headroom. Base the choice on peak memory and expected growth, not only the current average.
How to increase RAM on a DigitalOcean Droplet
DigitalOcean documents two CPU Droplet resize choices: one that increases CPU and RAM only, and another that increases CPU, RAM, and disk. Use the first when your existing disk allocation is sufficient; choose the second when storage must grow as well.
Resize procedure
- SSH to the Droplet during the maintenance window.
- Shut it down cleanly from the command line. DigitalOcean recommends this approach to help avoid data corruption. On a systemd-based Linux image,
sudo shutdown -h nowis a typical command; use the shutdown method appropriate to your distribution. - Confirm that the Droplet is powered off, then take a current snapshot from the control panel.
- Open the Droplet’s resize controls and choose either the CPU-and-RAM-only option or the CPU, RAM, and disk option.
- Select the larger plan and apply the resize. The platform may perform a migration before the Droplet becomes available again.
- Boot the Droplet after the operation completes.
- Reconnect over SSH and run
free -h. Confirm the new total memory, then check services, mounts, firewall rules, application health, and monitoring alerts.
After a DigitalOcean resize
Verify that the filesystem sees the expected disk size if you selected the disk-increasing option. Confirm database connectivity and run a representative request through the application. Keep the snapshot until the server has remained healthy through at least one normal traffic cycle.
Rank #2
- A-Tech RAM Memory compatible for select DDR4 Server and Workstation systems only; (*WILL NOT WORK with Desktop or Laptop Computers/PCs*)
- 16GB RAM Kit (2 x 8GB Modules); DDR4 DIMM 288 Pin; Speeds up to 2933MHz PC4-23400 (PC4-2933Y)
- ECC Unbuffered UDIMM; 1Rx8 - Single Rank x8; JEDEC DDR4 standard 1.2V
- Improves system performance, workload capacity, and reduces bottlenecks by increasing memory (RAM) resources
- Note: This memory is ECC Unbuffered and cannot be mixed with different ECC types such as ECC Registered, ECC Load Reduced, or Non-ECC Unbuffered; (Memory compatibility can vary among different system models and their installed components; please verify compatibility and follow memory channel guidelines to ensure maximum performance)
How to increase RAM on an Amazon Lightsail instance
Lightsail uses fixed bundles containing memory, vCPUs, storage, and a transfer allowance. You generally do not resize only the RAM field. Instead, create a snapshot of the existing instance and launch a new instance from that snapshot using a larger bundle.
Snapshot-and-replace procedure
- Open the Lightsail instance and create a current snapshot. Wait for the snapshot to finish before continuing.
- Open Snapshots in the Lightsail console.
- For the snapshot, choose Create new instance.
- Select a bundle with more memory. Review its vCPU, storage, transfer allowance, and recurring price, because all of those values are part of the fixed bundle.
- Create the new instance and wait for it to finish provisioning.
- Test SSH access, mounted storage, startup services, database connectivity, application responses, and scheduled jobs.
- Switch traffic to the new instance. Keep the old instance stopped or isolated until the new one is verified, then remove it only after confirming that the snapshot and backups are sufficient.
Preserve the public address
The default public IPv4 address changes when a Lightsail instance is stopped and started. If clients, DNS records, allowlists, or game-server listings depend on a stable address, use a static IPv4 address and remap it to the replacement instance after testing. Update DNS only after the new instance is ready, and account for resolver caching when estimating the cutover time.
How to increase RAM on a Vultr Cloud Compute instance
Vultr documents an upgrade through the instance’s plan settings. The Change Plan operation activates a plan with more vCPUs, RAM, and storage, and the documentation states that it does not change the instance data or file system.
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 →Change Plan procedure
- Open the instance in the Vultr control panel.
- Choose Settings, then Change Plan.
- Select a higher plan and review the new vCPU, RAM, storage, and billing details.
- Click Upgrade and confirm the change.
- Allow the platform to complete the upgrade. A reboot or short migration may still occur, so treat the operation as maintenance.
- Reconnect and verify memory with
free -h. Check application health, databases, mounts, firewall rules, IP reachability, and monitoring.
Provider differences that affect your cutover
| Comparison point | DigitalOcean Droplets | Amazon Lightsail | Vultr Cloud Compute |
|---|---|---|---|
| RAM-only choice | Offers a CPU-and-RAM-only resize, plus a combined CPU, RAM, and disk option. | Uses fixed bundles; memory is selected as part of the larger bundle. | Change Plan activates a higher plan with more vCPUs, RAM, and storage. |
| Change model | Resize the existing Droplet after shutting it down. | Create a new instance from a snapshot. | Change the instance plan from Settings. |
| Data and filesystem | Protect with a fresh snapshot before resizing. | Snapshot is the source for the replacement; test the new instance before retiring the old one. | Provider documentation says instance data and the file system are not changed. |
| IP considerations | Confirm address behavior in the console and your network design before maintenance. | Default public IPv4 changes after stop/start; a static IPv4 can be remapped. | Confirm address and migration behavior for the selected plan in the control panel. |
| Downtime expectation | Shutdown and possible migration or reboot. | Replacement and traffic cutover; DNS or static-IP work may be required. | Possible reboot or migration during the upgrade. |
Swap can prevent crashes, but it does not add VPS RAM
A swap file gives the operating system a place to offload inactive memory pages to disk. AWS describes it as room for the operating system to “offload inactive memory to disk.” Because disk is much slower than RAM, swap is a buffer for avoiding abrupt out-of-memory failures, not a substitute for a larger plan.
AWS guidance for Lightsail WordPress instances says swap can help reduce out-of-memory crashes on instances with less than about 1.5 GB of RAM. Its example creates a 650 MB swap file and persists it through /etc/fstab. Those figures are an example from that guidance, not a universal size for every workload.
When swap is useful
- A small instance experiences occasional memory spikes rather than constant pressure.
- A deployment, package installation, or backup briefly needs more working space.
- You need time to recover gracefully instead of having the kernel kill a database or application.
When swap is harmful
If the workload is continuously swapping, response times can become much worse and storage latency can increase. Monitor vmstat and application latency after enabling it. Do not describe a larger swap file as “more RAM” to users or monitoring systems: the VPS’s allocated memory remains unchanged.
Do these 3 things before closing this tab:
1Scan for outdated or missing drivers - takes under a minute2Clear out junk files and repair common Windows errors3Fix the driver behind crashes, sound loss and screen glitchesAfter a successful RAM upgrade, reassess emergency swap and worker limits. Remove or reduce temporary measures only after observing stable memory behavior, and preserve a reasonable safety buffer if the workload still has occasional spikes.
Rank #3
- A-Tech RAM Memory compatible for select DDR4 Server and Workstation systems only; (*WILL NOT WORK with Desktop or Laptop Computers/PCs*)
- 32GB RAM Kit (2 x 16GB Modules); DDR4 DIMM 288 Pin; Speeds up to 2400MHz PC4-19200 (PC4-2400T)
- ECC Unbuffered UDIMM; 2Rx8 - Dual Rank x8; JEDEC DDR4 standard 1.2V
- Improves system performance, workload capacity, and reduces bottlenecks by increasing memory (RAM) resources
- Note: This memory is ECC Unbuffered and cannot be mixed with different ECC types such as ECC Registered, ECC Load Reduced, or Non-ECC Unbuffered; (Memory compatibility can vary among different system models and their installed components; please verify compatibility and follow memory channel guidelines to ensure maximum performance)
RAM is not the same as storage
An attached block-storage volume increases disk capacity. It does not increase the memory available to processes. DigitalOcean describes volumes as network-attached block storage that can be attached or resized. Use a volume when the problem is full disks, log growth, media files, backups, or database data; resize the compute plan when the problem is exhausted RAM.
| Symptom | Likely action |
|---|---|
free -h shows low available memory, swap activity, or OOM-killer events. |
Investigate processes and limits, then move to a plan with more RAM if pressure is sustained. |
| The filesystem is full while memory remains available. | Clean up data or attach/resize block storage. |
| High I/O wait with adequate memory. | Investigate storage performance, queries, logs, and disk capacity. |
| CPU is continuously saturated. | Optimize the workload or choose a plan with more vCPUs; extra RAM alone may not help. |
Verify the server after the upgrade
- Run
free -hand compare total memory with the plan specification. - Run
vmstat 1 5and confirm that sustained swapping has stopped or reduced. - Check that all expected services are active and that no unit failed during boot.
- Test database connections, migrations, queues, scheduled tasks, uploads, and representative user requests.
- Confirm mounts and permissions, especially for attached volumes and application data directories.
- Check firewall rules, security groups, listening ports, TLS certificates, and remote administration access.
- Verify DNS, static-IP assignments, reverse proxies, and any allowlists that reference the server address.
- Review monitoring alerts for memory, CPU, disk, latency, and process restarts.
- Keep the pre-change snapshot or backup until the system has passed normal peak traffic.
Common failure modes and fixes
The new RAM is not visible
Confirm that the upgrade completed rather than remaining queued, and check the plan shown in the provider console. If the operating system still reports the old amount, reboot once if the provider requires it, then inspect boot and kernel logs. A container or control-panel limit can also hide host memory from the application.
The application still crashes
Read the kernel and application logs to identify the killed process. Check per-service limits, database connection counts, worker counts, and memory leaks. More RAM will not correct an application that allocates memory without releasing it.
The server is reachable by IP but not by hostname
Check whether the provider assigned a new address, then update DNS records, reverse-proxy targets, firewall allowlists, and external integrations. For Lightsail replacement workflows, remap a static IPv4 address when a stable endpoint is required.
The resize caused unexpected disk changes
Some plans couple additional RAM with additional storage. Compare the selected plan’s disk allocation with the previous one, and verify filesystem capacity and mounts before deleting the old snapshot or instance.
Performance is worse after enabling swap
Use vmstat to confirm active swapping. If swap-in and swap-out remain high, reduce memory demand or resize the plan; do not keep increasing swap while the application is constantly paging.
A practical decision checklist
- Is memory consistently low during real peak traffic?
- Do logs show out-of-memory kills or database termination?
- Have container, control-panel, runtime, and service limits been checked?
- Is the actual bottleneck CPU, storage I/O, database configuration, worker limits, or a leak instead?
- Does the chosen plan provide enough headroom for growth?
- Is a current snapshot or backup available and tested?
- Can the application tolerate the expected reboot, migration, or replacement?
- Will the public IP change, and have DNS or static-IP steps been prepared?
- Have memory, services, mounts, networking, databases, and monitoring been verified afterward?
Bottom line
Adding RAM to a VPS means upgrading its virtual-machine plan, not attaching storage or installing hardware. Diagnose memory pressure first, back up the instance, choose a plan with headroom, and follow the provider’s resize or snapshot-replacement workflow. Use swap only as a slower emergency buffer, then verify the operating system and every production service before declaring the upgrade complete.
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
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.
