0% found this document useful (0 votes)
17 views5 pages

Algorithmic Trading Platform

This document outlines the architecture of an Algorithmic Trading Platform (ATP), detailing its key components: the Market Data Adapter (MDA), Complex Events Processing (CEP) Engine, and Order Routing/Management System (OMS). It explains how these components interact to process market data, execute trades, and manage risks. The server is emphasized as the core of the ATP, designed for robust data processing and machine-to-machine interaction.

Uploaded by

choiburton
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
17 views5 pages

Algorithmic Trading Platform

This document outlines the architecture of an Algorithmic Trading Platform (ATP), detailing its key components: the Market Data Adapter (MDA), Complex Events Processing (CEP) Engine, and Order Routing/Management System (OMS). It explains how these components interact to process market data, execute trades, and manage risks. The server is emphasized as the core of the ATP, designed for robust data processing and machine-to-machine interaction.

Uploaded by

choiburton
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 5

Algorithmic Trading Platform

Table of Contents
Objectives ............................................................................................................................................... 2

Algorithmic Trading Platform (ATP) ........................................................................................................ 2

Server .................................................................................................................................................. 2

Market Data Adapter (MDA) ........................................................................................................... 3

Complex Events Processing (CEP) Engine ....................................................................................... 3

Order Routing/Management System (OMS) .................................................................................. 3

© Copyright QuantInsti Quantitative Learning Private Limited. 1


Objectives
The objective of this document is to teach you various components of the system
architecture of an algorithmic trading platform. You will understand how these components
work together to execute the trades.

Algorithmic Trading Platform (ATP)


Every trading system has two basic components within it. First, receiving data from
the exchange and second, placing orders in the exchange. In manual trading, we see market
data flickering on our screens and electronically or telephonically, we place orders for
execution. The analysis of data or decision making happens separately, either by tools and
software or by trader’s discretion. This decision making is the third part which is also built
within an algorithmic trading system.

Server
The core of an Algorithmic Trading Platform is built on a server. Before looking at the
parts, let us first understand what a server is:

A desktop computer is designed for human - machine interaction, and typically runs
on a user-friendly operating system with desktop applications that are used for lighter tasks.
A server is designed for machine to machine interaction. It interacts with the network and in
most cases might not have a graphical user interface (desktop screen). Server is generally
used for processing data 24x7 and is hence designed to be centrally manageable, robust and
reliable under load for years. An ATP has to deal with huge volumes of market data and it’s
processing which requires high specification and more robust machines. Hence, servers are
more suited for algorithmic trading platforms. The server interacts with the Application
Layer and the Trading Exchange/Broker. Application Layer is the graphical user interface for
the trading platform. It can be used for setting strategy parameters, viewing orders, trades
etc.

© Copyright QuantInsti Quantitative Learning Private Limited. 2


The three main parts of an algorithmic trading platform are explained below:

Market Data Adapter (MDA)

An adaptor is used to connect two different incompatible devices work together.


Here adaptor refers to that part of the ATP which helps to connect and interact with the
exchange or broker. Market Data Adaptor is used to receive data from exchange or vendor
and convert it to the format that the trading system can understand. Market Data may
come directly from the exchange or in some cases may be purchased from a vendor or
trading platform provider. In the same way orders can be directly sent to the exchange or
through a broker depending on the type of membership held with the exchange. Data
received from exchange is raw and unprocessed, consisting chunks of data packages, which
when processed by the adaptor gets updated and stored in the order book.

Complex Events Processing (CEP) Engine

CEP engine is the brain of the system. The main strategy lies here. This is where the
incoming market data is analysed, and the trading algorithm generates trading order. Maths
calc is connected to the CEP engine and helps in handling complex mathematical operations
in a dedicated calculations block (see Maths calc block in the diagram). It can perform
statistical tests, calculate options greeks, etc.

Order Routing/Management System (OMS)

Routing means directing the traffic in a certain direction in a network. In our case,
Order Routing refers to directing orders received from CEP engine to the desired exchange
or trading destination/broker after processing them. CEP sends instructions to OMS which
first passes the orders through a Risk Management System (RMS) that takes care of the risk
in real time. It may include checks for price, order quantity/value, etc. OMS then converts

© Copyright QuantInsti Quantitative Learning Private Limited. 3


the orders to a format that exchange understands. Exchanges need to interact with different
trading platforms and hence require the information regarding order. in a specific format to
be able to process them. Same goes when an exchange is sending information that is,
market data to the trading platforms. Different exchanges might have their own “Native”
formats for this interaction. Similarly, “FIX” is a standard format most exchanges use for
interacting with ATPs. In some cases exchanges have their own Native format and apply a
layer of FIX format above it to receive and send data In such cases sending data directly in
native format may make processing faster as it skips the extra layer. But using the
exchange’s native format might also involve more efforts in terms of maintenance, which
can become more complex when you are maintaining adaptors for multiple exchanges.

Main blocks in the system architecture of an algorithmic trading platform

© Copyright QuantInsti Quantitative Learning Private Limited. 4

You might also like