Looking at big data performance

My very first Intel blog entry!! Exciting!!

I just came back from a 3 days trip to San Jose where I attended the Hadoop Summit. It was great to see the level of energy around all big data technologies and how they address (and solve) real world problems. I must admit I spend too much time in our lab and don’t get the opportunity to talk often enough to the community. Well that was my chance. I listened and learned a lot from the Hadoop leaders and attendees.  I was amazed that most of the performance talks I attended filled up their rooms! People genuinely seem to care about speeding up their workloads!!?? What an exciting development! We (at Intel) do a lot of research on performance and my focus is Hadoop performance!! A lot of our findings often end up in the oubliettes, which is unfortunate as I now believe the community might find them valuable.  After that revelation, I decided to share some of our work on a blog. It will be a collection of Java, Hadoop, NoSQL, real-time analytics performance tuning and analysis that didn’t fit into a white paper or publication!

Next week, we will discuss tuning Java GC for HBase!

Stay tuned!!

Pour de plus amples informations sur les optimisations de compilation, consultez notre Avertissement concernant les optimisations.