جميع الكتب

مهني

1 Markets I: The Ecosystem and Exchange-Traded Marketsالأسواق عبر الإنترنت 2 Markets II: Rates, FX and Creditالأسواق عبر الإنترنت 3 Markets III: Commodities, Energy and Cryptoالأسواق عبر الإنترنت 4 Quantitative Methodsالأساليب عبر الإنترنت 5 Derivatives and Volatilityالمشتقات عبر الإنترنت 6 Rates, Credit, XVA and Riskالفائدة والائتمان والمخاطر عبر الإنترنت 7 Research Craft: Predictors, Backtests, Measurement, Portfoliosالبحث عبر الإنترنت 8 Strategies I: Equities and Futuresالاستراتيجيات عبر الإنترنت 9 Strategies II: Volatility, Relative Value, Macro and the Bank Desksالاستراتيجيات عبر الإنترنت 10 Microstructure and Executionالتنفيذ عبر الإنترنت 11 Market Making and High-Frequency Tradingصناعة السوق عبر الإنترنت 12 Machine Learning for Marketsتعلم الآلة عبر الإنترنت 13 Low-Latency Softwareالتكنولوجيا عبر الإنترنت 14 Networks, Hardware and Trading Infrastructureالتكنولوجيا عبر الإنترنت 15 Research, Data and Risk Platformsالتكنولوجيا عبر الإنترنت 16 The Desk and the Firmالشركة عبر الإنترنت 17 The Industry: Firms, Roles and Careersالمسارات المهنية عبر الإنترنت 18 The Interview Bookالمسارات المهنية عبر الإنترنت
التطبيقات حول المدرب تسجيل الدخول ابدأ القراءة

Quantitative Finance · المسرد

ما معنى Exactly-once processing؟

Definition 24.6 Research, Data and Risk Platforms · الفصل 24 — Messaging and Event-Driven Architecture

Exactly-once processing is the guarantee that each message’s effect on the consumer’s state happens once, whatever the crashes: obtained not from the transport, which can only deliver at least once, but by making the effect idempotent (Book 13, chapter 24) or by storing the state and the offset in one atomic transaction.

Fills lost and duplicated per 1 000 consumer crashes, over a hundred days of ten crashes in each mode. Committing first loses about half a batch per crash (50 323 per 1 000); committing after duplicates as much (48 652); state and offset in one transaction, none. Data: fig_eventlog.py.
Figure 24.2. Fills lost and duplicated per 1 000 consumer crashes, over a hundred days of ten crashes in each mode. Committing first loses about half a batch per crash (50 323 per 1 000); committing after duplicates as much (48 652); state and offset in one transaction, none. Data: fig_eventlog.py.
The largest end-of-day position error in any account and stock, for each of the hundred days, sorted. Ten crashes a day leave a median worst error of 3 500 shares when committing first and 3 800 when committing after, and up to 5 100 and 6 500; the atomic mode ends every day exact. Data: fig_eventlog.py.
Figure 24.3. The largest end-of-day position error in any account and stock, for each of the hundred days, sorted. Ten crashes a day leave a median worst error of 3 500 shares when committing first and 3 800 when committing after, and up to 5 100 and 6 500; the atomic mode ends every day exact. Data: fig_eventlog.py.
اقرأ في الفصل →