TrueNAS proves that the TLC premium isn't worth it for your next NAS SSD

Published Sep 26, 2026, 4:30 PM EDT His love of PCs and their components was born out of trying to squeeze every ounce of performance out of the family computer. Tinkering with his own build at age 10 turned into building PCs for friends and family, fostering a passion that would ultimately take shape as a career path. Besides being the first call for tech support for those close to him, Ty is a computer science student, with his focus being cloud computing and networking. He also competed in semi-pro Counter-Strike for 8 years, making him intimately familiar with everything to do with peripherals. Not all SSDs are created equal, and no spec embodies that more than the NAND type. With regard to NAS storage, most will tell you to always use TLC, and to avoid QLC NAND like the plague. TrueNAS just published a detailed case that the rule is wrong for a lot of people. In a post on how OpenZFS handles both NAND types, the company argues that QLC performs better than its spec sheet suggests, and that OpenZFS's write path is part of why. For the pool that holds your media, photos, and backups, I think TrueNAS makes a great case. QLC's weaknesses don't show up in conventional NAS workloads It's usually a fine performer Source: Kingston Technology The tradeoff between QLC and TLC NAND is very clear and well documented. TLC stores three bits per cell and QLC stores four, which gives QLC more density and a lower cost per terabyte in exchange for slower native writes, higher latency under sustained load, and fewer program/erase cycles. Nobody, TrueNAS included, thinks that these downsides don't exist. What TrueNAS disputes is where those costs show up, and they say QLC comes out ahead for capacity-bound, read-heavy, or bursty workloads like media archives, backup targets, and data that's written once and read often. That describes almost everything in a typical DIY bulk storage pool. This is your Jellyfin library, your Immich share, and your nightly backups. QLC does have a very clear write speed cliff though, and it's easy to go plummeting off of it in some NAS use cases, but most of the pain associated with QLC is when it's used in a normal PC environment. The endurance numbers also look a bit better than QLC's reputation suggests, at least for something that's being used as a bulk-storage pool. QLC drives carry lower endurance ratings than their TLC equivalents, but even so, consumer QLC ratings usually work out to a few hundred full-drive writes over the warranty period. A media library or backup target writes most of its data once and then mostly reads it back, and routine ZFS scrubs only read, making it a bit of a non-issue. ZFS shapes your writes before the drive sees them OpenZFS is the important part of this equation At a hardware level, QLC is generally adequate for NAS use, but the filesystem is where QLC can really start to be effective. The following also applies to any OpenZFS system, whether that's TrueNAS, Proxmox, or plain Linux, the benefit to QLC drives is the same. ZFS collects writes in memory as a transaction group and commits them to the pool as one larger, more sequential operation instead of many small, scattered ones. The default commit timeout is five seconds, but heavy write loads can trigger commits sooner, and slow devices can stretch them out. That batching of writes suits QLC well. A stream of small random writes keeps the drive refilling its pSLC cache and raises write amplification, while fewer, larger commits let the drive absorb a batch and move it to native cells before the next commit arrives. This also only applies to asynchonous writes, which doesn't matter for bulk pools, but would affect something like a database or virtualization over NFS. Your network is the real speed limit The networking runs out of road before the drives do The other hard truth about a NAS is that the drives typically don't hit their limit before the networking does. A typical gigabit NAS tops out around 116MB/s, and 2.5GbE SMB transfers land around 290MB/s. Any SSD can read faster than either, so QLC's read speed doesn't matter over the network. Writes are where the drive and network start to compete a bit. Many SATA QLC SSDs are rated for around 80MB/s after their cache fills at lower capacities like 1TB, but can go up to and beyond ~150MB/s at 2TB and up. On a gigabit connection, that's enough to keep up with the link speed. NVMe drives fair a lot better, usually far surpassing gigabit and 2.5 GbE speeds. This really only matters when you're writing tens of gigabytes continuously, so it's not worth worrying about, but it is worth at least keeping in mind. TLC will always be the better option TrueNAS accepts that, too We've established that QLC is adequate for home NAS use, but TLC is generally going to always be better in every department besides price. On its Pure Storage comparison page, TrueNAS themselves says it uses faster, higher-endurance TLC flash below 100TB and criticizes its competitor for charging premium prices for QLC. Its QLC post also concedes that sustained, write-heavy workloads still favor TLC, because that's where QLC's weaknesses can turn into measurable degradation over time. The differences in consumer versus enterprise drives matters as well. In their QLC post, they're referring to enterprise SSDs. Consumer QLC drives will commonly ship without a DRAM cache, which makes them significantly slower than an enterprise equivalent, and even slower than a TLC drive. QLC doesn't beat TLC, but it's better than you think TrueNAS isn't trying to prove that QLC beats TLC, nor does it. It's just illustrating that NAND type isn't as important in a NAS build than you might think, especially on anything using OpenZFS. For the pool holding your media and backups, ZFS smooths out the writes, your network caps the speed, and endurance ratings sit far beyond what write-once data will use. Spend the TLC money where writes actually happen, and let QLC hold everything else.