TruePut™

What your hard drive isn’t telling you.

Ever started an offload that says 40 minutes—only to return 15 minutes later and see the load bar is now at 1 hour and 20 minutes?

That disconnect between what’s promised and what is actually delivered highlights a deeper issue:peak performance numbers can’t predict how storage is used in production.

When we make an external storage purchase decision, performance metrics like “up to 2800MB/s!” are one of the top criteria we consider. But what customers often don’t realize is that poor cooling, small DRAM caches, and other factors mean those numbers rarely hold up.

What matters more is predictable,sustained throughput that doesn’t crumble halfway through your data transfer. At iodyne, we call it TruePut (see how we measure it). 

Let us explain more.

What is throughput?

Throughput is the amount of data that moves through a system over time—measured in Megabytes per second (MB/s) or Gigabytes per second (GB/s). This is the number you will find printed on the packaging of any hard drive or portable SSD. That’s the number widely advertised when a hot new, fastest-ever device comes out.

But there is a caveat to that.

The problem with peak performance

The throughput figure is typically measured at peak performance under ideal conditions. Most drives can perform at their advertised speed for small loads of data, or the first few seconds and minutes of a job. 

However, various benchmarks that industry leaders like Tom’s Hardware have been running show that throughput is hardly a metric a professional production can rely on. 

Why does performance drop?

A few reasons most systems fail to sustain speed:

🌡️ Thermal throttling

🧠 System or OS overhead

🔌 Cabling and interface bottlenecks

🔒 Encryption overhead

🧰 RAID complexity

💾 DRAM/SRAM caching and SLC caching

All of these can degrade the advertised throughput, just when you need it most.

What does this mean in a real-world scenario?

Ideal conditions can start falling apart at a film set under hot lights or on location, for a VFX studio rendering terabytes by the hour, or during a live event recording and broadcasting multi-cam footage. These are everyday conditions for media professionals, but not ideal for “throughput.”

For example, a commercial production team expects to ingest 8TB of raw footage in 1.5 hours at 1400MB/s. Except that speed only lasts for 30 seconds. The real transfer time takes nearly 3 hours, holding up the rest of the cast and crew from moving on to the next shot or wrapping the set. Multiply that by 30+ days of production, and you’ve lost 45+ hours just waiting on data.

A stalled ingest doesn’t just waste time —it burns money. With a few editors and assistants on the clock, every three-hour delay can cost thousands in project-based budgets. No one’s billing for that time, but everyone feels it: interrupted creative flow, late nights, and stalled schedules that block the next project from even starting. Multiply that across a season or a campaign, it adds up to a substantial opportunity cost.

It’s not bad planning. It’s that most storage can’t sustain its peak performance.