This blog is to share with the general public an internal milestone that was achieved by K7’s Product Engineering Team.
K7 Web Categorization, engine hosted on K7 Cloud Infrastructure, has been tirelessly serving our customers for close to a year. Recently we zoomed past an average serving up of more than one billion queries every month. Web Categorization is the process by which a website previously unknown to K7 will go through an automated Artificial Intelligence (Machine Learning) system that evaluates the page content, and thus predicts the category of website. This categorization of web pages is single-handedly responsible for providing the Category-Based Web Filtering for our enterprise customers. It also provides our home users with an enhanced browsing experience.
Those of you who are “cloud savvy” would probably be wondering if one billion queries is a low number considering our customer base – it indeed is. Utilizing efficient caching on the client side, as well as smart use of internet infrastructure & protocols, we are able to optimize the load on our cloud servers. The server software running on these cloud servers has been developed in-house using highly-optimized but traditional programming techniques to minimize hardware resources, and maximize throughput such that the inflow which peaked at 10000 queries per second a few months ago was handled with ease.
In this day and age of cloud computing the use of interpreted/JIT-compiled languages is predominant. However, there is still a special place for custom-built C/C++ compiled server software, if you care to extract every ounce of performance out of your hardware and provide a quality service to clients seamlessly.
Image courtesy of:
sendgrid.com
Product Engineering Team
If you wish to subscribe to our blog, please add the URL provided below to your blog reader: https://labs.k7computing.com/feed
