Conversion from Pebibits per second to Tebioctets per second

Information about the Pebibit per second (Pibit/s)

The pebibit per second is a unit of data transfer rate equal to 1,024 tebibits per second, which is two to the fiftieth power bits per second. Its symbol is Pibit/s. It is the binary counterpart of the petabit per second, and the two differ by 12.6 per cent.

No deployed system runs at this rate, and the unit therefore describes either an aggregate or a laboratory result. Optical transmission records set on single fibres reach a petabit per second, and the binary figure for the same experiment is 12.6 per cent lower — a difference that matters when comparing results between papers that use different conventions.

Where the unit is genuinely appropriate is in describing structures built from powers of two. The total switching capacity of a very large network fabric, built from ports and buffers that are all binary, is a binary quantity divided by time, and reporting it in decimal units discards the arithmetic that produced it. The same applies to the summed memory bandwidth of a machine whose channel count is a power of two.

In octets a pebibit per second is 140,737,488,355,328, which is 128 tebioctets per second. That is the storage of a thousand large consumer drives moved every second, and it exists only as a total across many thousands of parallel paths inside a single facility.

The size of the discrepancy at this level is the argument for the whole IEC series in miniature. What began as a harmless 2.4 per cent at the kibibit is now an eighth, and it compounds by 2.4 per cent at every further step. A convention that was acceptable for small numbers becomes untenable for large ones, and the point of the binary prefixes is to make the distinction visible before that happens.

For a converter, the arithmetic is unremarkable: multiply or divide by 1,024 the appropriate number of times, and by eight to reach octets. What matters is that the tool does not silently substitute the decimal unit when it sees a value it cannot label precisely.

One pebibit per second equals 1,024 tebibits per second, 140,737,488,355,328 octets per second, or about 1.126 petabits 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.