Adform's technology architect Peter Milne will share his insights about achieving High Load in advertising technology at QCon London 2017. Read the interview with Peter to learn more about his role and how Adform deals with the High Load.Read More
Quite a long time ago, we wrote about how we were joining streams in Apache Storm.
The ideas described there are still valid and being used. However, that was only the beginning—many new nuances have come up while join was running in production. Also, we wanted to overcome the limitations pointed out in the first article.Read More
The ACM SIGKDD Conference on Knowledge Discovery and Data Mining, or short KDD is one of the largest premier academic conference on data science and large scale data mining and machine learning. Many well-known researchers choose to publish their research at KDD and the conference attracts most of the big names in the machine learning community. The KDD Cup challenge is held in conjunction with the conference and every year it gathers many participants competing to win the award.
As data scientists, we here at Adform are fortunate to have a vast wealth of complex real-time bidding display advertising data ready and waiting to be analyzed. Such data offers fresh and interesting research possibilities to academics and, luckily for us, Jun and his Ph.D. student, Weinan Zhang, were more than happy to apply their technical skills to our data.Read More
I was going to start with a comparison to Java "generics" and C++ templates just to show that .NET is "better". But I decided not to do that because we already know that .NET is wonderful. Don't we? So let's leave it as a statement.Read More
For us, it all has started with a date… The date of our wedding. A group of product managers and owners came to us, developers, with an idea of the Cross-Device solution and “some” requirements.Read More
Our story has begun with a "simple" need to join data. It seemed easy, as all of us have done joins in databases before, but there was a catch.
The catch was related to the type and amount of data we have.Read More