While doing some optimization testing, I discovered different result sets when running the same lookup request multiple times. My examples show requests with different DatapointsPerSend, but the same problem exists even when the DPS value is held static. I also tested other symbols and had the same problems. I am running the current beta version and all lookups are done via the TCP/IP API. This is a very repeatable problem with a sample size of 5 to 10 requests. I found that I usually end up with 3 or 4 different record count sets. Of the sets that have the same number of rows, I have seen discrepancies in those as well. It appears to me that this data is being calculated, as opposed to being static. I have log files if needed. Here are some examples:

Two Same History Requests
HTD,SLV,120,0,,,1,18,25000
HTD,SLV,120,0,,,1,18,20000

Two Different Results
***** HTD,SLV,120,0,,,1,18,25000
18,2012-05-14 09:31:27,27.6300,300,522559,27.6300,27.6400,5940,0,0,C,
18,2012-05-14 09:31:27,27.6300,200,522759,27.6300,27.6400,5941,0,0,C,
18,2012-05-14 09:31:27,27.6300,696,523455,27.6300,27.6400,5942,0,0,C,
18,2012-05-14 09:31:27,27.6300,100,523555,27.6300,27.6400,5943,0,0,C,
18,2012-05-14 09:31:27,27.6300,500,524055,27.6300,27.6400,5944,0,0,C,
18,2012-05-14 09:31:27,27.6300,510,524565,27.6200,27.6300,5945,0,0,C,
***** HTD,SLV,120,0,,,1,18,20000
18,2012-05-14 09:31:27,27.6300,300,522559,27.6300,27.6400,5940,0,0,C,
18,2012-05-14 09:31:27,27.6300,200,522759,27.6200,27.6300,5941,0,0,C,
18,2012-05-14 09:31:27,27.6300,696,523455,27.6200,27.6300,5942,0,0,C,
18,2012-05-14 09:31:27,27.6300,100,523555,27.6200,27.6300,5943,0,0,C,
18,2012-05-14 09:31:27,27.6300,500,524055,27.6200,27.6300,5944,0,0,C,
18,2012-05-14 09:31:27,27.6300,510,524565,27.6200,27.6300,5945,0,0,C,
*****

***** HTD,SLV,120,0,,,1,18,25000
18,2012-05-14 14:06:30,27.6100,100,10498253,27.6000,27.6100,118621,0,0,C,
18,2012-05-14 14:06:30,27.6100,200,10498453,27.6100,27.6200,118622,0,0,C,
18,2012-05-14 14:06:30,27.6100,492,10498945,27.6100,27.6200,118623,0,0,C,
***** HTD,SLV,120,0,,,1,18,20000
18,2012-05-14 14:06:30,27.6100,100,10498253,27.6000,27.6100,118621,0,0,C,
18,2012-05-14 14:06:30,27.6100,200,10498453,27.6000,27.6100,118622,0,0,C,
18,2012-05-14 14:06:30,27.6100,492,10498945,27.6100,27.6200,118623,0,0,C,
*****

Thank you.

We receive the data in two different feeds from the exchange and then combine and interweave the data. One feed has the trades and the other has the bids and asks. The reason some of the data is slightly different between runs is that some of the runs may be from different server farms and they didn’t combine/interweave it at exactly at the same point (moment in time) on each system so they are just so slightly off from each other. Having duplicate systems is very good, but it’s also very difficult to make them all perfectly the same with the huge volumes of data that go though them.

---

Visit our new online forums! DTN employees actively monitor the forums to provide you the highest level of support in the industry. Our forums also has an announcements area where you can learn about new features or changes within DTN. You can even use the forums to exchange trading ideas or tips and tricks on how you use your DTN data to be more profitable.

Go to http://forums.dtniq.com now to start sharing information with fellow DTN subscribers!