Mebibyte to Exabyte

MB (megabyte) = 1,000,000 bytes (SI decimal). MiB (mebibyte) = 1,048,576 bytes (IEC binary). MiB is 4.86% larger. This gap is why a file manager on Linux showing "512 MiB" of free RAM and a marketing sheet showing "512 MB" of RAM are technically different: the marketing sheet refers to fewer bytes.

The original CD-ROM standard defined capacity as 74 minutes of audio or 650,000,000 bytes. Technically this is 650 MB in SI terms, or approximately 620 MiB (since 650,000,000 ÷ 1,048,576 ≈ 620). However, the CD industry loosely used "MB" to mean 650 × 10⁶ bytes. Some media used 700 MB (≈ 668 MiB). This inconsistency is a classic example of the pre-IEC ambiguity.

Docker reports image sizes in decimal MB (e.g., "docker images" shows 150 MB), but the underlying layer storage on disk uses binary-aligned block sizes. A "150 MB" Docker image actually occupies roughly 143 MiB on disk before compression. Compressed layers further complicate things: a 150 MB uncompressed image might only transfer 50 MB over the network. Container registries like Docker Hub display compressed sizes, while "docker images" shows uncompressed — leading to frequent confusion in CI/CD pipeline size budgets.

One mebibyte (1,048,576 bytes) holds about: one minute of MP3 audio at 128 kbps (≈ 960 kB, so slightly under 1 MiB), a medium-resolution JPEG photo (0.5–2 MiB), about 200 pages of plain text, or the complete text of a short novel. A typical Linux kernel image at boot is 8–12 MiB compressed.

RAM chips are physically organized as binary address grids — each address line doubles the capacity, so sizes must be exact powers of 2 (4 GiB = 2³² bytes, 8 GiB = 2³³ bytes). USB flash drives use NAND flash that is also binary internally, but manufacturers reserve variable amounts for wear levelling, bad block management, and controller firmware. A "64 GB" USB drive might have 64 GiB of raw NAND but only expose 59.6 GiB (64 × 10⁹ ÷ 2³⁰) to the user — the label uses decimal marketing, unlike RAM which honestly reflects binary sizing.

One exabyte = 1,000,000 terabytes = 1,000 petabytes. If you filled 1 TB external hard drives and stacked them end to end, 1 EB worth would stretch roughly 200 km. In content terms: 1 EB can store about 250,000 years of HD video, or about 100 billion hours of music at 128 kbps. All the data produced by the Large Hadron Collider per year is about 15 petabytes — still 67× less than one exabyte.

Global data creation, capture, copy, and consumption is estimated at roughly 2.5 exabytes per day (IDC 2023 estimate), growing roughly 23% annually. This includes IoT sensor readings, financial transactions, social media posts, surveillance camera footage, scientific instrument output, and all other digital activity. Most of this data is transient and never stored long-term.

Amazon Web Services, Microsoft Azure, and Google Cloud each store estimated tens to hundreds of exabytes of customer data in their cloud platforms. Meta (Facebook/Instagram) stores an estimated 100+ exabytes across all data types. The NSA's Utah Data Center is estimated to hold yottabytes in capability, though actual stored volumes are classified. Collectively, global cloud storage is in the hundreds-of-exabytes range.

An exabyte (EB) = 10¹⁸ bytes (SI decimal). An exbibyte (EiB) = 2⁶⁰ bytes = 1,152,921,504,606,846,976 bytes — about 15.3% larger. This is the largest practically relevant gap between SI and IEC units in storage contexts. For a data center procuring 10 EB of storage, the SI vs IEC difference represents about 1.5 EB of capacity discrepancy in the contract.

Data archaeology is the practice of recovering information from obsolete storage media and formats — 9-track magnetic tapes, 8-inch floppy disks, MiniDiscs, Zip drives, and early optical formats. The challenge is threefold: hardware to read the media no longer exists or is failing, file formats and encoding schemes are undocumented, and magnetic media degrade over time (tape has a 10–30 year shelf life). At exabyte scale, organisations like national archives face the prospect of vast digital collections becoming unreadable within decades. Active migration strategies — periodically copying data to current formats and media — are the only reliable defense, but the cost scales linearly with data volume.