Gigabyte to Exbibit

Hard drive manufacturers measure 1 TB as 1,000,000,000,000 bytes (decimal). Windows displays storage in gibibytes (binary) but historically labelled them as "GB" — so 1,000,000,000,000 bytes ÷ 1,073,741,824 ≈ 931 GiB, which Windows displayed as "931 GB". macOS (since 10.6) correctly reports the same drive as "1 TB" using decimal GB. The drive is not lying; the OS was using a binary unit with a decimal label.

8 GB RAM is the current minimum for gaming; 16 GB is the recommended standard for most modern games at 1080p and 1440p; 32 GB benefits heavily multitasking systems or games with large open worlds. Memory-intensive tasks like video editing, 3D rendering, and running large language models locally typically require 32–64 GB or more.

A 4K movie in H.264 or H.265 encoding is typically 50–100 GB on Blu-ray; streaming services compress aggressively to 15–25 GB for 4K HDR content. Netflix's 4K streams average about 7 GB per hour; the downloaded version via the Netflix app for offline viewing is roughly 3–6 GB per hour at high quality settings.

1 GB of mobile data supports roughly: 2–3 hours of music streaming, 1 hour of HD video streaming, 2–3 hours of web browsing, or 30–60 minutes of video calling. Social media apps with autoplay video are heavy consumers — TikTok and Instagram Reels can use 300–600 MB per hour of active use.

AI model sizes vary enormously. GPT-2 (2019) is about 1.5 GB; Llama 2 7B is roughly 13 GB in float16 precision; Llama 2 70B is about 130 GB. GPT-4-class models are estimated at 500+ GB. Quantised (compressed) versions are smaller: a 4-bit quantised 7B model fits in about 4 GB, runnable on a modern laptop. Training requires far more — the training dataset, gradients, and optimizer states for a 70B model can occupy 1–2 TB of GPU memory across a cluster. The trend toward larger models is driving consumer GPU memory from 8 GB to 16–24 GB as a baseline for local AI inference.

An exabit (Ebit) = 10¹⁸ bits (SI decimal). An exbibit (Eibit) = 2⁶⁰ bits ≈ 1.1529 × 10¹⁸ bits (IEC binary). Exbibit is 15.29% larger — the cumulative product of using 1,024 instead of 1,000 at each of six prefix steps. This is the largest practically relevant SI vs IEC gap for bit units in current storage contexts.

Exbibit is used in: computer science academic literature on exascale computing, theoretical storage system design papers, and formal IEC/IEEE standards. No commercial product, OS, or consumer application currently displays exbibits. It is primarily a unit for academic and standards consistency — ensuring the IEC prefix family extends uniformly from kibi- to exbi- (and beyond to zebi- and yobi-).

After exbibit (Eibit, 2⁶⁰ bits) come: zebibit (Zibit, 2⁷⁰ bits) and yobibit (Yibit, 2⁸⁰ bits). These are defined in the IEC 80000-13 standard but have no current practical applications. The IEC binary prefix family deliberately mirrors the SI prefix family, ensuring consistent naming as computing scale continues to grow.

Frontier (Oak Ridge, 2022) achieved 1.194 exaFLOPS, with its Slingshot-11 fabric moving data at aggregate rates measurable in exbibits per second across 9,408 nodes. Aurora (Argonne, 2024) targets similar throughput with over 63,000 GPUs. At these scales, a single checkpoint of a full-system simulation can exceed 1 Eibit of state data, making exbibit a natural unit for describing I/O bandwidth requirements.

The IEC currently defines up to yobibit (Yibit, 2⁸⁰ bits). In 2022, the SI system added ronna- (10²⁷) and quetta- (10³⁰), but the IEC has not yet created matching binary prefixes (ronnibit? quettibit?). With global data creation projected to exceed 1 yottabit annually by the 2030s, pressure is mounting for the IEC to extend the binary prefix family — though the naming convention ("ronnibi-"?) remains an open question.