Home Forums Discussions General Discussion Data Sharing and Speed

We know AC 'shares' data with their partners and about a year ago we were given the option in the User Agreement to opt out of sharing our data, I opted out immediately and received confirmation that I was opted out.
So here is my question and I hope the responses from those more programming savvy can address it:
Once opted out I didn't receive a 'new AC' and no one from AC asked to remote in to alter my AC so I presume whatever data collection mechanism/process built into AC remains active. Presumably wherever the data goes there is a 'don't use data from me' switch turned on.
So what are the processes within AC to collect what data? which interactions/tab clicks activate those processes? what possible delays can these processes cause? could it be that until said collection process completes the program is put on hold until released by said process?
Yes it's 4AM and the free association part of my brain is spinning but would appreciate some learned responses.

No, it doesn't work like that. AC has access to your program and databases just as Microsoft knows what updates you need or how your A/V programs knows when your definitions are outdated. Software reminds you when you should upgrade.

AC is simply data mining to retrieve certain information that they then share with so that your 500 transactions become 500,000 transactions and then number crunchers and analysts can deduce certain patterns from this information. So, maybe the query is to find out how many 35 to 42 year olds on xyz drug have xyz outcome. They can then predict or make assumptions based on this. Google does this, Micrososoft does this.

They are not taking your data, then marking it as not usable. They are retrieving, hopefully, unidentifiable data on those practices who did not opt out. This can be done in many different ways. But, it is almost certain this is not done in real time. It is done at certain times and likely done rather quickly.

The performance degradation would be negligible. You could use something like Wireshark to see what information is going back and forth, but given this is happening sporadically you would need very sophisticated logging and filtering.

So, the answer is no. This is not slowing your system down. If it were, then I would be seeing degradation on 6.1.2 which is super fast and fairly bug free except for one annoying thing. And, I should add, I am staying right there.

Just woke up so not a very clear answer. Sorry.

Thanks for the reply, yes I've downloaded Wireshark and have to figure out how to set up the filters as it gives a lot of info if not filtered out.
So then for those practices opted in for data sharing is it assumed while those docs are home comfortably tucked away in bed at night AC (or one of their partners) is getting into their computers database and mining away?

What gets me confused is that when I run the speed test on my client computers I get 0.979s usually less than 1s but this does not tally up with the actual speed of amazing charts which I find slow. This is using v8.2.4 on a hard wired network connection. (was getting 1.5s on same set up on V7.1.3) Not being an expert this would tell me there is other than just speed of the network causing the slowness of amazing charts.

The speed test is simply sending one to four packets (infitestimanlly small) compared with reading and writing data to a database. It is just a quick method to compare one computer to the next.

Make yourself a large file, hopefully a GB, but a 100MB file and just copy and paste it from your computer to your "server," and note the time. This will give you a better idea.

My speeds are generally < 0.7 all on all computers, but it has no indicator of my actual speeds with any version. Now, if I used wireless and the speed on that wireless connection were 5.2, that would give me some info to use overall.

So, it is going to boil down to coding of the program, which will relate to cachiing of data on AC and how AC reads and writes to the database. As some have said earlier, the two biggest performance enhancers will be having a wired network (by far) and the fastest processor possible as it appears AC is only using one core of the possibly four your processor has such as in an i7 processor. The latter part of this was figured out with the help of two other people on the board who should receive credit, but I will keep them namesless for now in case they do not wish to have their names written here.

Bottom line, your tests were a good idea overall, but do not bear out your conclusion.

Sql express can use up to 4 cores guess AC not programmed to use all 4?

It's how your processor works with each program. It has nothing to do with SQL. So, when doing some testing, e.g. opening a chart, opening eRx, starting AC, we would take note of how high and in what cores the spikes were. We saw this on two test stations. (You can tweak SQL to run processor cores in parallel, but I certainly wouldn't recommend it on a production system based on this small sample size.

I have looked at others, and this doesn't seem to be the same. For more valid and reliable information, I would contact Guardian Sandeep.

Bert's right about the file size coming into play. A few hundred megabits can be transferred during initial sign in and opening of charts. The AC test is basically testing the time it takes to access the database and read back a value. Think of the database like series of Excel spreadsheets. The time you see is to get to one spreadsheet a read a value. Now when AC opens a chart, especially in the new version, it's accessing multiple tables (multiple spreadsheets) in this example. One sheet has the all the names, the other has all the diagnoses. So the factor in play here is that you have to access too many tables when you open a single chart. If it takes 1 second to get to one table, then another second to get to the next... it just compounds the problem.