How BotFoundry models multi-instance MT4/MT5, Freqtrade, and headless Playwright /dev/shm memory math from official software docs and public VPS tariffs. (BotFoundry Last Updated: November 12, 2026)
BotFoundry is an independent technical sizing workbook and unattended daemon hardening publication maintained by the BotFoundry Editorial Desk. Our editorial team ("BotFoundry Editorial") builds browser-based memory calculators for algorithmic traders running MetaTrader 4/5, crypto engineers deploying Freqtrade or Hummingbot, and automation teams scaling Playwright or n8n Queue Mode.
We fund BotFoundry without display ad networks, paywalls, or email lead-capture gates: when readers click outbound /go/ links to provision hosting at Vultr, UltaHost, Hostinger, InterServer, or ChemiCloud, we may earn a referral commission at zero extra cost to the buyer.
We model every hardware estimate in src/sizer-engine.mjs deterministically from official upstream documentation (MetaQuotes terminal64.exe portable guides, Microsoft Playwright Docker --shm-size=2g notes, Linux cgroup v2 MemoryMax manuals, and Java 21 ZGC specifications) paired with public vendor plan tables retrieved on 2026-11-12.
Specifically, our 4-layer memory equation sums OS base footprint (0.65 GB Ubuntu 24.04 systemd vs 2.80 GB Windows Server 2022 RDP), shared runtime overhead (Docker/Wine64/Redis pools), active per-instance heap scaled by chart or DOM density (0.36 GB to 0.78 GB per worker), and a 20%–30% volatility buffer.
We never publish synthetic ping charts as hands-on benchmarks. Wherever multi-day packet captures are still pending — specifically direct host-level telemetry verification on your target instance — BotFoundry holds those /compare/ pages as unindexed reference until verifiable raw logs are archived.
Readers can audit every public vendor rate cited on BotFoundry directly against official checkout pages and report formula updates through our editorial contact channel.