Marketfeed 2000 Beta Users: Handling A Data Firehose
THIS WEEK'S LEAD STORIES
Lines are in at several beta sites for Reuters's high speed data feed product, tentatively dubbed "Marketfeed 2000." Among the sites already connected are Drexel Burnham Lambert, Shearson Lehman, Merrill Lynch, Morgan Stanley, and Fidelity Investments in Boston.
Marketfeed 2000 -- Reuters's answer to the "Give us the data and get lost" crowd -- is being positioned as a live market data source for the analytic engines of trading research groups. But before rocket scientists can access the data, users must decide how to pre-process the tidal wave of information.
The data feed offers tick-by-tick updates for upwards of 100,000 instruments in logical record format. More than 200 updates/second will be delivered during peak traffic periods, each requiring an average of 40 bytes of data. The data feed is compressed at the Reuters host using Huffmann compression, and decompressed at the customer's premises.
Building Blocks
Each Marketfeed 2000 installation is served by two data feed transmitters at the nearest Reuters data center, two fully redundant 56 kilobits/second leased lines, and a digital switching unit. Beyond these basic building blocks, however, the decision tree for users branches out.
Rapid access to Marketfeed 2000 data requires that all or part of the database be cached by the data feed server. If you want access to the entire database, then you need nearly 100 megabytes of RAM. The question is, who's going to provide the cache?
By next year, Reuters plans to make available a data server capable of caching the entire feed. The server is a variant of the the recently announced 80386-based Intel iRMX 320 series servers specially equipped with four 32 megabyte memory boards. The server was developed by Rich Inc.
For Marketfeed 2000 enthusiasts not willing to wait that long, Reuters will be offering a non-caching server within the next four weeks. The non-caching server has just completed three months of exhaustive testing at the Hauppauge development facility.
The "pass-through" server package includes the software required for data feed decompression and protocol conversion, but not the code needed to address the cache, reportedly the most complex part of the user application.
Eventually, tool-kits will be available for customers wishing to cache the data feed and develop applications on systems running either VAX VMS or Unix and C. Preparations are also being made to feed IBM mainframes via LU 6.2.
Tortoises And Hares
The speed and the level of commitment exhibited at the various beta sites varies wildly, as more ambitious players shoulder more of the development burden themselves in hopes of gaining a strategic edge.
That edge comes at a price, however, as users are trading off the risks and the costs of internal development against the convenience and reliability of a packaged solution.
Some of the beta sites accepted the data feed months ago and have been building their software from scratch. Others seem to be waiting to see what Reuters comes up with. Some are doing both.
Drexel Burnham Lambert, for example, is now accepting a reduced subset of the data feed, about 2000 stocks, through an AT server for development purposes. Sources at Drexel say the firm will use the data decompression and protocol conversion software provided by Reuters, but remains undecided on the final server selection.
In the meantime, cache address code is being written at Drexel to support a Stratus minicomputer, but the Intel server developed by Rich is also under consideration.
Users who want the fault-tolerance of a Stratus server without the headache of writing their own code are being offered a twin Intel server configuration by Reuters.
The Marketfeed 2000 marketing team at Reuters is committed to building maximum flexibility into its product line, but can would-be users afford to wait?
Only users who have a paid subscription or are part of a corporate subscription are able to print or copy content.
To access these options, along with all other subscription benefits, please contact info@waterstechnology.com or view our subscription options here: http://subscriptions.waterstechnology.com/subscribe
You are currently unable to print this content. Please contact info@waterstechnology.com to find out more.
You are currently unable to copy this content. Please contact info@waterstechnology.com to find out more.
Copyright Infopro Digital Limited. All rights reserved.
As outlined in our terms and conditions, https://www.infopro-digital.com/terms-and-conditions/subscriptions/ (point 2.4), printing is limited to a single copy.
If you would like to purchase additional rights please email info@waterstechnology.com
Copyright Infopro Digital Limited. All rights reserved.
You may share this content using our article tools. As outlined in our terms and conditions, https://www.infopro-digital.com/terms-and-conditions/subscriptions/ (clause 2.4), an Authorised User may only make one copy of the materials for their own personal use. You must also comply with the restrictions in clause 2.5.
If you would like to purchase additional rights please email info@waterstechnology.com
More on Data Management
New working group to create open framework for managing rising market data costs
Substantive Research is putting together a working group of market data-consuming firms with the aim of crafting quantitative metrics for market data cost avoidance.
Off-channel messaging (and regulators) still a massive headache for banks
Waters Wrap: Anthony wonders why US regulators are waging a war using fines, while European regulators have chosen a less draconian path.
Back to basics: Data management woes continue for the buy side
Data management platform Fencore helps investment managers resolve symptoms of not having a central data layer.
‘Feature, not a bug’: Bloomberg makes the case for Figi
Bloomberg created the Figi identifier, but ceded all its rights to the Object Management Group 10 years ago. Here, Bloomberg’s Richard Robinson and Steve Meizanis write to dispel what they believe to be misconceptions about Figi and the FDTA.
SS&C builds data mesh to unite acquired platforms
The vendor is using GenAI and APIs as part of the ongoing project.
Aussie asset managers struggle to meet ‘bank-like’ collateral, margin obligations
New margin and collateral requirements imposed by UMR and its regulator, Apra, are forcing buy-side firms to find tools to help.
Where have all the exchange platform providers gone?
The IMD Wrap: Running an exchange is a profitable business. The margins on market data sales alone can be staggering. And since every exchange needs a reliable and efficient exchange technology stack, Max asks why more vendors aren’t diving into this space.
Reading the bones: Citi, BNY, Morgan Stanley invest in AI, alt data, & private markets
Investment arms at large US banks are taken with emerging technologies such as generative AI, alternative and unstructured data, and private markets as they look to partner with, acquire, and invest in leading startups.