Conversion from Gibioctets per second to Tebioctets per second

Information about the Gibioctet per second (Gio/s)

The gibioctet per second is a unit of data transfer rate equal to 1,073,741,824 octets per second, which is 1,024 mebioctets per second. Its symbol is Gio/s. It is the unit of memory bandwidth and of the fastest storage interfaces, and the binary counterpart of the gigaoctet per second, from which it differs by 7.4 per cent.

Memory is where the unit belongs most naturally. A memory channel transfers a fixed number of octets per clock cycle, and that number is a power of two, so the resulting bandwidth is a binary multiple of the clock frequency. A machine with several channels reaches tens of gibioctets per second, and an accelerator with stacked memory reaches thousands.

Storage has caught up. A fast solid-state drive on the current interface sustains several gibioctets per second, which means that for the first time the drive and the memory are within an order of magnitude of each other. That convergence has changed how software is written: the old assumption that reading from disc is thousands of times slower than reading from memory no longer holds.

The unit appears in benchmark output, in system monitoring displays and in the specifications of processor interconnects. All of these count in binary because the structures they measure are binary, and reporting the result with a decimal prefix would introduce a seven per cent error for the sake of a familiar-looking label.

For a sense of what the rate means, one gibioctet per second copies a two-gigaoctet film in under two seconds and fills a one-teraoctet drive in about a quarter of an hour. Anything at this speed is faster than every external connection in an ordinary building, so the limiting factor moves inside the machine.

The distinction from the decimal unit matters most in procurement and capacity planning. A specification that requires 10 gigaoctets per second and a system that delivers 10 gibioctets per second are not the same, and the difference of 7.4 per cent is the sort of margin that decides whether a design meets its requirement.

One gibioctet per second equals 1,073,741,824 octets per second, 1,024 mebioctets per second, or about 1.074 gigaoctets per second.

Information about the Tebioctet per second (Tio/s)

The tebioctet per second is a unit of data transfer rate equal to 1,024 gibioctets per second, or two to the fortieth power octets per second. Its symbol is Tio/s. It is the binary counterpart of the teraoctet per second, and the two differ by 10 per cent.

Nothing outside a large machine moves data this quickly. The rate describes the memory bandwidth of an accelerator with stacked memory, the internal fabric of a high-end processor package, or the aggregate throughput of a parallel filesystem spread across thousands of drives. All of these are built from binary structures, and their totals are binary quantities divided by time.

High-performance computing is where the unit is written most often. A supercomputer's storage system is specified by how many tebioctets per second it can deliver to the compute nodes, because that number determines how quickly a simulation can save its state and resume. A machine that computes quickly but writes slowly spends its time waiting.

The ten per cent difference from the decimal unit is significant in that context. A filesystem procured to deliver 10 teraoctets per second and one delivering 10 tebioctets per second differ by a whole teraoctet per second, which in a facility of that size represents a substantial fraction of the hardware budget.

In bits a tebioctet per second is 8 tebibits per second, and in decimal terms about 1.1 teraoctets per second. Expressing the same rate four different ways is routine at this level, because the storage industry, the memory industry, the network industry and the standards bodies each prefer a different one.

For everyday comparison, a tebioctet per second would fill a large consumer hard drive in about twenty seconds. No external interface carries this; the figure describes movement between components inside a single system, where the wires are short and there are very many of them running in parallel.

Graphics processors have brought the rate within reach of a single component. A stack of high-bandwidth memory bonded directly to the processor die delivers well over a tebioctet per second to the chip that uses it, and a card carrying several such stacks passes a few. That bandwidth, rather than raw arithmetic speed, is what limits the training of large models: the arithmetic units sit idle unless the memory can keep them fed. The same reasoning explains why supercomputer designers spend as much effort on the paths between memory and processor as on the processors themselves, and why the rate is quoted in binary units when the memory it describes is addressed in powers of two.

One tebioctet per second equals 1,024 gibioctets per second, 1,099,511,627,776 octets per second, or about 1.100 teraoctets per second.