Tag Archives: database applications

Chicago July PUG with Bruce Momjian

I have just one thing to say – it was a great meetup! I was worried, that the even was scheduled not for the best day: it’s the end of July, when people are either heading to vacation, or at least just trying to spend as much time outdoors, as possible. Besides, for some reason nobody likes to meet on Thursdays.

In spite all of the above – the attendance was great, and the audience was really engaged. Do I need to mention that Bruce is the greatest speaker in the Postgres world?!

The presentation was brilliant,  over an hour just passed by, and nobody even asked whether we can have a break 🙂

And now I really hope  this won’t be the last time Bruce is visiting Chicago and talking at PUG!


Leave a comment

Filed under events, People

The joy of team work!

Last week I had to work a lot! I mean, A LOT! But work-wise that was one of the happiest weeks, and I wanted to share with everybody why it felt so happy.

Those of you who worked with me before know, that my favorite activity is working with  applications, because optimizing for applications is way more difficult and way more fun than optimizing reports. Yes, you can impress everybody, including yourself, reducing the report run time from one hour to one minute. But how much cooler is it to reduce the page load time from 30 sec to 0.1 sec?!  Especially when you have a power not only to write the best queries ever, but also to design the data model “the right way”.

When you do the application database work, the most critical part is to work in close contact with the  application developers. And depending on what kind of team you are lucky or unlucky to have, it may be the best or the worst part of your professional experience.

My IT team here in Braviant is one of the best I ever worked with, which was proved one more time last week. The most difficult part has always been connecting the db work and the app work, like: I’ve selected all this data for you, can you read it from the output I am providing? Or: we can give you all input parameters that way, can you process them? Our app developers have already made a huge step “in my direction” agreeing not to use ORM, but to read the output of the database function. Next step – we hit the  wall exactly where I expected. I’ve spent a half of Saturday writing my code, so that the app developers could start using in Monday morning… and now they are saying they can’t process correctly the embedded record sets! I’ve heard that many times before, and each time in a couple of hours I would hear: Hettie, there is no way! Let’s do it “the old way”, we know how… That time, however my team kept trying to find a solution, and watching these efforts made me to start thinking how I can change the output on my side. After several iterations going back and forward, we came up with a pretty neat way to return the records, which could be used right away, and even a better approach, which, however would require more work from me, and could not be done on the spot.

And you know – I totally understand, when people hate rewriting one piece of code multiple times, which makes me appreciate even more the willingness to rewrite later, when I will come up with the more automated solution from my side…

I have already written a lot in this post, and I am not sure whether it all make sense, but let me try to summarize.  I loved that everybody were willing to compromise, to make adjustments, that there were no “just because” statements, that the whole team was focused on the goal to build the application right from the very beginning, so that we won’t need to worry about performance six months down the road.

Hope it will continue that way.  Except of me working on Saturdays part 🙂


Leave a comment

Filed under People, Systems, Team and teamwork

I am going to be a panelist at the ChicagoSC event

This post is mostly for my Chicago colleagues, but who knows? May be somebody from nearby could also come.

Please check out this event: Panel: ORM’s and Data Management. I will be one of for panelists, and the topic of ORM and ORIM is near and dear to my heart, as all of you know! So I am very excited about this event and honored to be a part of this panel. Please try to  attend :)!

Leave a comment

Filed under events, Systems, talks

Finally many -many months later – our presentation at PG Open 2014

Leave a comment

Filed under Systems

My and Grant’s PG Open presentation

View it here

Leave a comment

Filed under talks

Sometimes people just do not know what are they writing about…

Some time ago my manager shared with me a couple of links about “current state of data and databases”. One of them turned to be a perfect illustration to how we get where we got. This article The definitive guide to the modern database actually describes the current status of the databases, and what it out there.

One paragraph caught my attention:

This is the case when database engine and application are scaled-in together on the operating system level, so both of them use the same data without transmitting it through proxies. Also, a benefit of such approach is the increasing ease of use — monotonic code to deal with wrappers is no longer needed, and for some users this can become a more decisive factor than performance.

The funny thing is, that this is EXACTLY what led to our performance problems in my company!

Granted, the author never experienced how bad an OO code “can” become


Filed under Data management, Systems

My future talk at PG Open 2014

This past week my talk proposal for PG Open 2014 was accepted, so now I actually have to write it :). It’s interesting, how everything got changed in course of the past year. A year ago I was just trying to return back to the world of conferences and presentations, and my talk acceptance for PG Open 2013 was a huge milestone for me.

And this year, three conferences later, dozen of talks later, it seems like “something expected” (although my submission to PG Conf was not accepted).

It has being a great year so far, I’ve learned a lot about myself and other people, I’ve learned my strengths, learned not to be afraid to speak up and to take responsibility; I learned to compromise more than ever, and to work towards reaching a mutually acceptable solution – as long as it takes.  A year ago I won’t imagine myself being where I am now….

In my future PG Open talk I am going to officially introduce the term HDAT, and in contrast to the last year I will be talking more about the database aspect of it. Specifically, I want to answer multiple questions I’ve being getting on how to reuse the database code designed for optimal performance. How to factor it without actually factoring it. And I hope that Grant will help me with the case study:)


Filed under talks