Evaluating Black Scholes Millions Of Times A second on a Macbook Air with MDBX

MDBX is a tick data processing system. It pioneered concepts such as RTQUERY and is built from first-principles to empower real-time financial workloads.

The problem of, in real time, updating the value of a derivative based on a change in the underlying is rather common in financial workloads. MDBX excels at it. The following is a performance test of (synthetic) SP500 top of book updates being fed to calculate the real time quotes of an option book for the instrument. While the bleeding edge in option pricing has long moved past vanilla black scholes, many pricing methods are a derivative of it.

(All numbers are valid as of 2026.07.20 on a consumer Macbook Air with auxiliary workloads. As always, if you are interested in enabling new workloads with MDBX email ron@modolap.com.)

The MDBX RTQUERY UOPSX program:

(program
  (output stream)
  (backend rt)
  (scan {options} :as options)
  (scan {quotes} :as quotes)
  (asof-join options quotes :as joined
    (on t t) (by ticker ticker) (prefix quote_))
  (project joined :as book
    (field contract contract)
    (field ticker ticker)
    (field quote_t quote_t)
    (field sent_ns quote_sent_ns)
    (field sequence quote_sequence)
    (field spot (/ (+ quote_bid quote_ask) 2.0))
    (field theoretical
      (black_scholes_price
        (/ (+ quote_bid quote_ask) 2.0)
        strike years volatility rate is_call)))
  (root book))

The following chart isn’t merely a tuned, point-in-time measurement but in some sense a scaling law; MDBX scales throughput of output with input in a nearly linear fashion, MDBX is hyper-elastic.

Empower Your Financial Workloads

MDBX is available today, supporting both managed, BYOC, and on-prem installations. It can replace either generic systems not tailored for financial use cases, or partial in-house solution. There exist the first order effect: faster, cheaper, more scaleable, better experience - Yet also the second order — enabling real time transformation with the proprietary RTQUERY, and discovering use cases considered inviable beforehand. Email me at ron@modolap.com.