4
$\begingroup$

I have been looking for a low cost solution to effectively store and query tick and bar data.

Databases like kdb+ and Streambase are too expensive for me.

Building a custom solution with SSAS (Sql Server Analysis Services) is one of the options that can think off. Has anyone attempted this?

If you did attempt this, how did you organize your data? Any ideas in that direction would be helpful.

Improve this question
$\endgroup$
4
  • 2
    $\begingroup$ Bars and Tick data are very different by nature. Bars data actually allow you to zoom out of tick data. If kdb+ is too expensive for you, then I assume you're not doing this for a firm, so how do you plan exploiting tick data with a small budget? $\endgroup$
    – SRKX
    Jul 31, 2012 at 9:15
  • $\begingroup$ SSAS allows for good aggregations of data, Tick data could be aggregated very well using SSAS. I am doing this for a small firm in India. kdb+ is out of their budget. $\endgroup$
    – nunespascal
    Jul 31, 2012 at 10:04
  • $\begingroup$ I assume you meant Vertica as an example of column-oriented storage; Streambase is a CEP. $\endgroup$
    – chrisaycock
    Jul 31, 2012 at 11:18
  • $\begingroup$ Yes, I meant it as an example of column-oriented storage. $\endgroup$
    – nunespascal
    Jul 31, 2012 at 11:23

1 Answer 1

Reset to default
1
$\begingroup$

As you said, this sort of financial data can be well aggregated using cubes; icCube is for example a fast in-memory OLAP server you can access via XMLA clients or JAVA or Javascript native API (on top of the MDX language and HTTP protocol).

[edit after comment] open/close/high/low are types of aggregation supported by icCube; so creating this kind of 'bars' feature in icCube is a matter of defining several measures (with their own aggregation type) from the tick data (lowest granularity and only one to be persisted). Then around these measures several dimensions can be built (time, financial instrument types, etc...). Quite a classic star schema actually with no specific requirements.

Improve this answer
$\endgroup$
1
  • $\begingroup$ Thanks for your reply. Any recommendations on how you organized the data? $\endgroup$
    – nunespascal
    Sep 24, 2012 at 3:55

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.