2011. 8. 11. 10:42 Market Data

TickBase

TickBase is a real-time database server for super-high-speed capture, storage and retrieval of financial market data.

Capture Real-Time Data and Gain Control

Market feeds provide traders with enormous volumes of real-time information. Yet traditional databases simply aren't fast enough to capture and store this avalanche of tick data, leaving most firms unable to fully utilize the tick data they already pay for. Leading Market Technologies has developed a solution to this problem - TickBase.

TickBase provides super-high-speed, real-time capture, storage and retrieval of tick data from a wide range of sources, architectures, external feeds and client internal sources. This server captures market data as it comes down the wire, stores that data, and then makes it available for immediate or subsequent analysis. You can use TickBase to:

  • Filter out consecutive, redundant, and outlier tick values from instruments showing huge volumes of activity but little movement.
  • Save space using Open-High-Low-Close (OHLC) format for instruments that you are only peripherally monitoring.
  • Capture important news headlines and internal messages.
  • Capture your firm's own trades.
Because TickBase achieves a more accurate and manageable database, reduces CPU load, and keeps you informed, TickBase places you in control.


TickBase Benefits

High speed

Traders and analysts demand access to the most recent data, especially during periods of high-volume trading and significant market changes. TickBase can capture very high rates of data -- enough to handle the most demanding market volumes. In contrast, commonly used SQL and other relational databases can get swamped in periods of peak market activity. Only TickBase can deliver the information you need -- now.

Faster data access

TickBase is efficient. It can be used to capture data on over 50,000 different instruments simultaneously on a dedicated mid-range UNIX server. TickBase serves up intra-day tick data to client application "on demand" -- on any instrument, without limit and without the time delays of going to external sources. A typical TickBase configuration will serve up one megabyte of tick history in one second. People use TickBase to help make crucial decisions faster.

Scalable growth 

The TickBase is parallel and scalable. You can process a virtually limitless amount of information by daisy-chaining TickBase servers together.

Data protection

TickBase is deployed on a server, not at the client application level, so the data reservoir remains safeguarded from power outages and other accidents occurring on end-users' machines.

Flexibility

LMT designed TickBase to provide fully open data access. Use pre-installed interfaces for the most popular real-time data distribution platforms or create a custom interface using the open TickBase API to link TickBase to any data source.

Easy access

Access TickBase using MS Excel, LMT's EXPO, TickView, or use TickBase's flexible API to bring data into the analytical package of your choice as well as your own applications.



TickBase Features

Disk Handling

TickBase employs a proprietary architecture and data storage strategy for speed and space efficiency. TickBase makes it possible for you to write to many disk partitions simultaneously, creating scalable databases distributed over many hard drives on a single machine.

Filters

TickBase can be easily configured to filter out bad or missing data. Among TickBase's filter types are:

  • intervals between ticks
  • context-sensitive percentage change between ticks
  • absolute value (high and/or low)
  • removal of identical values

Interfaces

LMT designed TickBase as an open client/server solution. It integrates easily with in-house proprietary systems as well as off-the-shelf software. TickBase provides pre-installed interfaces to:

  • Reuters TRIARCH SSL
  • Tibco TIB
An open, portable, C source code API allows you to create a custom interface to any data source such as:
  • Tick-by-tick market feeds
  • Internally created information
  • News wires
  • Trades and inventory
  • Client transaction

Database cleansing

The TickBase database administrator facility clean up "dirty" data upon command. For example, define an outlier value, and TickBase will remove all ticks greater than, less than, or equal to the specified value for a specified time frames.

Database merging

You can merge data from multiple databases into TickBase easily because TickBase recognizes a wide range of ASCII-type data file formats. Just define a record type and modify the API to deliver that database to TickBase for storage. If you choose, you can specify a file or list of date/time/ value/ records for merging into TickBase.

Automatic database transfers

Transfer data to or from TickBase automatically by setting TickBase to call another process at periodic intervals.

Customization 

If you wish to modify TickBase, this can be done by taking advantage of its programming interface in source code form.

Warehousing Market Data 

TickBase lets you specify the number of days after which you will archive data, on an instrument-by-instrument basis.

Standard Storage Options for Tick Data 

Using TickBase you can store market and trade data as raw ticks, bars of arbitrary period, the closing value of the day, the CHLO of the day, and/or an average value for the period. You can also conduct "one-shot" queries to acquire a tick value at a given time or list of times during the day.

Platform Availability

TickBase currently runs on the following:

  • SUN Solaris
  • HP PA-RISC 9000 Series running HP-UX

 

'Market Data' 카테고리의 다른 글

TIP. Wallstyouth, May 25, 2007  (0) 2011.08.25
Tick(Time Series) Database Discussion  (0) 2011.08.12
TREP-VA for Quantitative Trading  (0) 2011.08.11
TREP-VA for TCA  (0) 2011.08.11
TREP-VA for Compliance  (0) 2011.08.11
Posted by karlsen

블로그 이미지
Pricing, hedging, risk-managing a complex derivative product
karlsen

태그목록

공지사항

Yesterday
Today
Total

달력

 « |  » 2024.7
1 2 3 4 5 6
7 8 9 10 11 12 13
14 15 16 17 18 19 20
21 22 23 24 25 26 27
28 29 30 31

최근에 올라온 글

최근에 달린 댓글

최근에 받은 트랙백

글 보관함