Human Engineering AG

data communication and contribution software

  • Increase font size
  • Default font size
  • Decrease font size

PICOS Introduction / Overview

The system allows concurrent distribution of financial information input to all the well-known "public" services (Reuters, Telekurs, Bridge, ADP, Bloomberg, others) and customer specific services (Host, Servers and other data stores) at the same time.

Possible input sources are


  • Reuters Triarch ( SSL )


  • TCP/IP
  • XML data stream ( Java Input Tool )
  • XML data stream ( Custom Application )
  • Web Browser Input ( HTML)

Serial Port

  • Reuters Marketlink ( Reuters DCS , input from trading system )
  • Reuters RMF / CI ( input from trading system )
  • Bridge R4 ( input from trading system )
  • Reuters Insertlink Third Party Contribution( input from Reuters Insert Terminal )

Windows DLL

  • Direct calls from within MS-Excel. This is a preferred solution, if no trading system is used.


  • We provide an API ( C / Java ) to allow input from within applications

Supported output protocols and services

  • ReutersMarketlink , RMF/C1
  • BloombergMarketlink , Multi Protocol Feed ( ask for details )
  • BridgeESIP , R4 , Marketlink
  • TelekursTelekurs Contributor Data Input
  • Internet, HTML , XML , WML (WAP)
  • customer specificspecial protocols , host access , data base …

Reuters Marketlink Contributing Protocol

We support the update of small pages , big pages and records . We process the return messages from Reuters and start actions (throttling) and/or forward them to the PICOS/CMS Control Center.

On throttling warnings from Reuters, updates to the corresponding page/record are not forwarded for the time recommended by Reuters (if configured).

The output to Reuters is highly optimized

  • Eliminating of duplicate data
  • Character update for page fields
  • Put overload in a cache not in a queue
  • There can be more then one line to Reuters for load sharing

Modularity and Flexibility

The modular design of PICOS results in total flexibility and high resilience.

The usual limitation upon throughput is the serial line to the service provider (throttling and flow control). PICOS is able to handle and control a pool of lines to a service provider thus dramatically increasing throughput. This results in rates quoted by the Information Provider more accurately representing the currently quoted price and increases the probability of your contributed rates being used in composite service provider string records.

Universal Platform

PICOS has been designed to be independent of a trading system architecture, and will therefore run on a variety of hardware platforms such as Sun Solaris 2.3 to 2.6 inclusive, Sun Sparc hardware or Windows NT 3.5x / NT4.0 Intel CPU. This flexibility ensures that PICOS can be implemented at a low cost and fits into the business hardware and operating environment strategy.

No Single Point of Failure

The system has been designed so that it can be built to have no single point of failure and has no limitations upon the geographical distribution of components. The Central Management module allows the system administrator to control, manage and configure the distributed system. Management of all processes via SNMP is available as an extra cost option. Central or distributed control of contributed prices is therefore available to the user.

This modular design enables the optimum use of processing power and offers the potential of using highly resilient distributed systems. The system can however run on one single CPU if reliability and resilience is not a prime requirement.

An additional benefit of the modular approach is that a new service provider can easily be integrated into the system.

Differences and Enhancements to other Multicontributing Systems

  • Customer’s choice of input sources.
  • Extensive range of output services.
  • Total flexibility to integrate new services and meet changing customer requirements.
  • Flexibility to change the presentation of data for different services.
  • Dynamic configuration / reconfiguration.
  • Software solution based on UNIX and / or Windows NT operation system.
  • Distributed architecture gives optimum performance and resilience.
  • Parallel transmission to service lines allows automatic load balancing to take place.
  • Priority transmission to service lines ensures that urgent updates are not interrupted.
  • Scalable architecture allows a system to be built that will cope with many 10s of thousands of updates per day; the largest user is currently contributing over 700,000 updates per day to internal and external sources.
  • Modular design allows a fault tolerant and resilient system to be built.
  • Dramatic increase in contribution throughput.