Do these 3 things before closing this tab:
1Fix the driver behind crashes, sound loss and screen glitches2Repair Windows errors before they cause bigger problems3Scan for outdated or missing drivers - takes under a minuteSome links on this page are affiliate links: if you buy through them we may earn a commission, at no extra cost to you.
The best TrueNAS server is not the one with the fastest processor or the biggest cache SSD. It is the one whose drives, memory, controller, network, cooling and backup plan fit its workload. ServeTheHome’s genuine TrueNAS CORE hardware guide remains useful as a historical framework, but it was last updated June 4, 2020. Treat its product suggestions as dated, not as a 2026 shopping list.
Before buying parts, decide whether CORE is the right operating system for a new build. The official CORE documentation is for the 13.0 release family; current TrueNAS documentation also covers newer Community Edition/SCALE hardware. Existing CORE systems may have good reasons to stay put, while new general-purpose builds should compare both branches before choosing a platform. See the CORE documentation and the current SCALE hardware guide.
Then size the server around what it will do: basic file sharing, media and backups, virtual machines or databases, or high-throughput all-flash storage. For most builds, reliable drives, the right ZFS layout, sufficient ECC memory, a properly configured HBA, airflow, power protection and tested backups matter more than optional cache devices.
Free tools Windows power users keep installed
One-click scans. No signup required.
Start with the workload, not the parts list
TrueNAS CORE can run on a wide range of x86-64 hardware, but “it boots” is not the same as “it is a dependable storage server.” Count the drives you need now and later; estimate usable capacity; identify whether clients mostly read, write, or access many small files; and decide whether the server will also run applications, jails, or virtual machines. Those answers drive the CPU, memory, vdev layout, HBA and network choices.
- Basic home file server: prioritize dependable drives, an ECC-capable platform if practical, sufficient RAM, simple networking and backups.
- Media and backup server: size capacity and redundancy first. A faster network may help multiple clients, but it does not make a single hard disk faster.
- Virtualization, iSCSI or database workloads: plan for more RAM, CPU capacity, low-latency storage and predictable synchronous-write behavior.
- 10/25GbE or all-flash: check that the pool, HBA, PCIe topology, CPU and network can all sustain the target workload. A fast NIC alone cannot do it.
TrueNAS CORE’s official hardware guide lists a two-core x86-64 processor, 8 GB RAM, a 16 GB SSD boot device and two identically sized devices for a single pool as baseline guidance. These are not universal production recommendations. The newer TrueNAS hardware guide gives a 20 GB SSD boot baseline for its newer releases; do not silently apply that figure to CORE 13.0. Consult the guide for the branch you actually install: CORE hardware requirements and newer TrueNAS hardware guidance.
CPU and motherboard: buy a balanced platform
Ordinary SMB or NFS file serving rarely calls for a high-end desktop CPU. Processor demand rises with encryption, compression, many concurrent clients, iSCSI, virtual machines, plugins or applications, transcoding, deduplication and high-speed networking. For transcoding, choose the CPU or GPU for the media workload; ZFS itself does not require a workstation-class processor.
Choose the CPU and motherboard together. Verify that the exact combination supports ECC operation if ECC matters to your build; a feature listed as “ECC compatible” does not by itself prove that error correction is enabled. Also check supported DIMMs, maximum memory, PCIe lanes and slot wiring, onboard networking, fan control, firmware support and idle power. IPMI or equivalent remote management is useful for a server that will live out of reach, but consumer boards often omit it.
Look closely at the board’s lane and port-sharing notes. Populating an M.2 slot can disable SATA ports; a second PCIe card can make the HBA or NIC run at fewer lanes than expected. Small boards may not have enough expansion for both a disk controller and 10GbE. Check the complete populated configuration—not just the number of slots printed in a product listing—and confirm physical clearance for the HBA and its cooling.
ECC memory and capacity
ECC memory is a strong preference for important data and systems expected to run continuously. It can detect and correct certain memory errors, reducing one route by which bad data might be handled in memory. It does not prevent disk, controller or software failures, accidental deletion, ransomware, fire or theft, and it does not replace backups. ECC requires a compatible CPU, motherboard, chipset and memory; test memory before deployment and watch system hardware logs.
CORE’s 8 GB figure is a basic-operation floor, not a sensible target for every server. The official guide suggests 8 GB for basic use with up to eight drives and roughly 1 GB additional RAM per drive beyond eight for many use cases. Treat these as planning guidance, not a sizing law. A practical starting range might be 8–16 GB ECC for simple file sharing, 16–32 GB for several users and routine services, 32–64 GB for larger pools or multiple services, and 64 GB or more for VMs, iSCSI or databases. Measure the actual workload and leave room for the operating system and applications.
RAM helps ZFS cache and manage metadata, but more memory does not automatically make every pool faster or overcome a slow vdev layout. Deduplication is particularly memory-hungry: TrueNAS documentation cites about 5 GB RAM per TB of storage as a planning guideline. Model the specific workload before enabling it rather than treating that ratio as a promise. L2ARC also uses RAM for metadata, so adding an SSD cache can make a memory-constrained system worse.
Boot media: use an SSD and keep a recovery path
For a new CORE build, use an SSD rather than a spinning disk or an ordinary USB flash drive as the system boot device. The CORE guide specifies a 16 GB SSD baseline and discourages USB sticks and hard disks for this role. A mirrored boot device can reduce downtime if one device fails, but it does not protect the data pool and is not a substitute for saving configuration backups.
The boot pool holds the operating system and boot environments; it is separate from the data pool. Save configuration backups regularly and keep a copy off the server. If a boot device fails, reinstall the same or a compatible release and restore the configuration. Boot environments can help roll back a system update, but they are not a backup of user data. See the official boot-environment guide and installation guide.
Data drives: match the media to the job
Drives are the core of a storage server. Select them for capacity, expected duty cycle, vibration environment, warranty, replacement availability and workload—not just advertised interface speed. NAS and enterprise HDDs are commonly better suited than desktop models to always-on multi-drive arrays. Before purchase, confirm:
Rank #3
- Performance-Oriented and Quiet Hardware Design: 32GB ECC RAM | 8-Core 2.2GHz Intel Atom CPU | 12x 3.5” Hot-Swap SATA Drive Bays | 2x RJ45 10Gigabit Ethernet LAN ports | Remote Management (IPMI) | 2x USB 2.0 Ports - 1x USB 3.0 Port | 1x Internal Boot Device | Built-in RAID | Boost performance by adding SSDs for read and write caching.
- Ideal for file-sharing, backup, multimedia processing, transcoding, and distribution, video surveillance, edge/remote office, development, personal cloud, and other small/home office & SMB applications. Broaden your Mini’s capabilities with VMs and an extensive suite of software plugins.
- TrueNAS software supports Windows, MacOS, Linux, and Unix clients and syncs with AWS, Azure, Dropbox and more. Supports NFS, SMB, AFP, iSCSI and S3 file sharing protocols. Use TrueCommand to manage multiple TrueNAS systems from a single interface.
- Includes Short Rail Kit - 19" to 26.6" rackmount depth for short racks and optional rubber feet for desktop.
- Item Weight: 41.7 lbs
- Recording technology: check CMR versus SMR. SMR drives can be a poor fit for RAIDZ workloads with sustained writes or resilvering.
- Interface: SATA is generally adequate for home and small-office arrays. SAS can make sense for enterprise shelves, dual-porting or expanders, but adds cost and compatibility checks.
- SSD behavior: for write-intensive roles, check endurance and power-loss protection rather than relying on a consumer label or peak benchmark.
- Compatibility: verify sector format, firmware behavior, cabling, backplane support, temperature limits and warranty terms.
- Capacity plan: mismatched sizes can leave capacity unused in some layouts. Buy with the intended vdev geometry and future expansion path in mind.
A 12 Gb/s SAS link does not make a mechanical disk deliver 12 Gb/s of sustained data. Link rate is not the same as media throughput, and pool performance depends on layout, workload and the number of drives working together.
Choose the ZFS vdev layout deliberately
There is no universally best layout. The right choice balances usable capacity, fault tolerance, random I/O, drive count, rebuild exposure and the way you expect to grow the pool.
- Mirrors: use pairs of drives with redundancy. Mirrors often provide better random I/O and can be easier to expand incrementally by adding another mirror vdev, at the cost of usable capacity.
- RAIDZ1: tolerates one drive failure in a vdev. With large drives or critical data, many buyers prefer more fault tolerance because a resilver can leave the vdev exposed to another failure.
- RAIDZ2: tolerates two drive failures in a vdev and is a common general-purpose choice for multi-drive arrays.
- RAIDZ3: tolerates three drive failures in a vdev and may suit very large arrays or higher-risk rebuild environments, with a further capacity trade-off.
- Stripe: offers no redundancy. A failed member can lose the pool, so this is generally unsuitable for important data.
Redundancy is not backup. A pool can still be lost to deletion, malware, fire, theft, administrator error or failures beyond its chosen tolerance. Keep independent backups and test restores. Plan the vdevs before buying drives: the capacity and expansion characteristics of a pool are not interchangeable with a conventional hardware RAID set.
HBA, backplane and cabling
ZFS should normally see individual disks directly. Use an HBA that exposes drives in IT/JBOD mode rather than putting the pool behind a traditional hardware RAID abstraction. TrueNAS identifies Broadcom/Avago/LSI SAS HBAs as common controllers, but no brand name guarantees that a particular card, firmware or configuration is right. Verify the actual model, supported firmware, IT mode, connector type, SAS generation, PCIe link and compatibility with the drives and backplane.
An HBA presents attached disks to the operating system. A hardware RAID controller instead manages disks behind its own RAID logic. A SAS expander can increase the number of attached drives, subject to bandwidth and compatibility; it is not the same thing as a SATA port multiplier. Check the full path from drive to backplane to cable to controller.
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
Rank #4
- Boost read performance on your TrueNAS Mini with a dedicated 480GB Read Cache (L2ARC) device.
- Take advantage TrueNAS's advanced algorithms to queue up your most frequently and most recently used data in high performance flash media.
- Hardware tested and qualified by the developers of TrueNAS.
- Confirm the card is genuine and has appropriate firmware; used or misidentified cards may not match their listings.
- Do not confuse RAID firmware with IT mode, and do not flash firmware without verifying the exact card, revision and procedure.
- Match cables and connectors to the card and backplane. A SAS drive will not work through arbitrary SATA-only hardware.
- Check PCIe lane allocation so the HBA has adequate bandwidth alongside the NIC.
- Provide airflow over the HBA. These cards can overheat in cases designed without active cooling across the heatsink.
Legacy component lists can be helpful for terminology, but avoid buying a specific old HBA solely because it appeared in a 2020 recommendation. Confirm that its firmware and release compatibility, condition, connectors and price make sense today.
L2ARC: add it only for a measured read-cache need
L2ARC is a secondary read cache, not a general-purpose SSD accelerator. It may help when a frequently reused read working set is larger than RAM, the workload is read-heavy, and the pool and network can benefit from cached reads. It does not replace RAM, speed every workload or guarantee better performance.
The CORE guide gives a rough L2ARC capacity guideline of 5–20 times system RAM, but that is not a target to fill automatically. Each cached block needs metadata in memory, so a large L2ARC can consume RAM. Before spending on one, check whether adding RAM, changing the pool layout or addressing the real bottleneck would help more. The practical rule is simple: buy L2ARC only after measurements show a repeatable read-cache limitation and the system has adequate memory.
SLOG: not a universal write cache
ZIL is the ZFS intent log; a separate log device used for synchronous-write records is called a SLOG. It primarily matters when applications issue synchronous writes—for example, some NFS, database, virtualization and enterprise workloads. It is not a write cache for ordinary asynchronous writes and many home NAS users do not need a separate device.
Quick wins for a faster PC:
Fix the driver behind crashes, sound loss and screen glitchesFind Drivers →Repair Windows errors before they cause bigger problemsFix Now →If the workload justifies a SLOG, choose a low-latency, endurance-rated device with power-loss protection and suitable platform support. An ordinary consumer NVMe drive without power-loss protection is a poor default for this role. Consider failure behavior and redundancy for the intended workload; a SLOG is part of the write path, not an expendable accelerator. Do not assume a legacy Optane model is the right modern purchase simply because older guides used one.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.Networking: buy the whole path
1GbE is enough for many home NAS uses, particularly when hard drives and ordinary file-sharing workloads are the limit. 2.5GbE can be a modest step up; 10GbE or 25GbE may suit multiple clients, faster pools or demanding transfers, but the server, switch, client adapters and cabling or transceivers all need to support the plan. A fast NIC does not make a single hard drive or slow pool deliver line rate.
Check that the NIC is supported by the specific TrueNAS/FreeBSD release and that the motherboard supplies enough PCIe lanes. SFP+ and RJ45 differ in switch compatibility, cabling, transceiver options, power and heat. Link aggregation can increase aggregate capacity across clients, but does not necessarily double the speed of one file transfer. Jumbo frames are optional; enable them only when every relevant endpoint and switch is configured consistently. See the CORE networking documentation.
Chassis, cooling, PSU and UPS
The case and power system are reliability components, not finishing touches. Choose enough drive bays for the planned pool and growth, and verify whether a backplane supports the required SATA/SAS drives and direct-attached or expander topology. Make drives serviceable and label them so you can identify the right device during a replacement.
Provide airflow over HDDs and the HBA, manage dust, and consider fan noise and replacement availability. Check power-supply quality, efficiency and headroom for simultaneous disk spin-up. Use only the correct modular cables for that exact PSU; modular PSU cables are not universally interchangeable. Avoid overloaded or poor-quality power connectors, and confirm the case leaves room for HBA cabling and cooling.
A UPS can give the system time to shut down safely during an outage or brownout, but it is not a backup. Configure and test USB or network signaling rather than assuming it works. Verify the complete failure path: interrupt power, confirm UPS communication, observe a safe shutdown before the battery is depleted, then confirm the server starts cleanly again. Pure sine-wave output may matter for some power-supply and load combinations.
Bare metal or virtualized?
A bare-metal installation is generally simpler to troubleshoot. Virtualizing TrueNAS can be appropriate for experienced administrators, but it adds a failure surface: the guest needs adequate memory and reliable networking, and disks should be presented individually through carefully configured controller passthrough or another supported direct-disk arrangement. Do not put ZFS behind a virtual hardware RAID abstraction. CORE’s installation documentation specifies at least 8 GB RAM for a TrueNAS VM, but real requirements increase with the pool and guest workload. See the CORE installation guide.
Which TrueNAS branch should a new build use?
CORE is FreeBSD-based and remains relevant for existing deployments, FreeBSD compatibility and workflows that specifically depend on its jails or plugins. Its official documentation is centered on CORE 13.0. Current TrueNAS documentation also provides hardware guidance for newer Community Edition/SCALE releases, whose direction and ecosystem differ; do not assume every CORE recommendation transfers perfectly.
| Situation | Reasonable starting point |
|---|---|
| Existing stable CORE server | Stay on CORE unless a concrete compatibility, feature or support need motivates migration. |
| New general-purpose NAS | Compare current TrueNAS Community Edition/SCALE hardware and software requirements before buying. |
| FreeBSD jail or CORE-specific workflow | CORE may remain the appropriate fit; verify required features and hardware support. |
| Linux containers or current app ecosystem | Give current Community Edition/SCALE priority in the evaluation. |
| Business system requiring vendor escalation | Assess official appliances and support options as well as a DIY build. |
Do not treat a branch choice as a casual in-place component decision. Check the target release’s migration guidance, application and hardware compatibility, and recovery plan before moving a working system.
Quick Recap
Common mistakes to avoid
- Putting ZFS behind hardware RAID or leaving an HBA in the wrong firmware mode.
- Assuming ECC works because a listing says “ECC compatible,” or treating non-ECC as equivalent.
- Using a USB stick as the only boot device in a continuously operating server.
- Buying L2ARC or SLOG before proving the workload needs it.
- Using a consumer SSD without power-loss protection for synchronous-write logging.
- Overlooking SMR behavior, sector-format differences, firmware quirks or replacement availability.
- Ignoring M.2/SATA sharing, PCIe lane limits, HBA cooling or backplane compatibility.
- Buying 10GbE without a matching switch, client, cabling and pool capable of using it.
- Mistaking RAIDZ redundancy for a backup, or failing to plan how the pool will grow.
- Skipping UPS shutdown tests, configuration backups or restore tests.
Final buying checklist
- Choose the TrueNAS branch and verify its current hardware guidance.
- Define drive count, capacity, workload, fault tolerance and future expansion before purchasing media.
- Confirm CPU, board and DIMMs support ECC operation if required.
- Size RAM for services and workload, not just the minimum install figure.
- Use SSD boot media and keep off-system configuration backups.
- Check drive recording technology, sector format, warranty and replacement availability.
- Verify the HBA model, firmware, IT mode, cables, backplane, PCIe lanes and cooling.
- Make sure the case, PSU and airflow support the drive count and simultaneous spin-up load.
- Buy networking as an end-to-end path, not a NIC in isolation.
- Test UPS signaling, shutdown, restart and backup restoration before trusting the server.
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.
