John Hart, Distinguished Engineer, Scalyr talks with John Furrier at Sclayr HQ in San Mateo, CA #theCUBE #Scalyr @SiliconANGLE theCUBE https://siliconangle.com/2019/06/06/speed-to-truth-scalyr-enables-search-complex-data-logs-at-warp-speed-scalyr/ Powered by thousands of servers Scalyr addresses this issue by essentially creating a way to use the massive compute power of a centralized system to serve its customers. Each customer gains access to a log management engine that is equipped to handle queries on a scale far greater than anything that might be available in-house. It’s a model very similar to what happens when anyone around the world queries Google. “Literally thousands of servers in that tenth of a second that Google is processing that query, 3,000 servers on the Google side may have been involved,” Newman explained. “Those aren’t your 3,000 servers, because you’re sharing those with 50 million other people in your data center region. But for a millisecond, those 3,000 servers are all for you.” This massive use of compute power represents a form of brute force to dramatically speed up the time needed to sort through log data and obtain an accurate and satisfactory answer to resolve a system issue. This kind of log management system must be flexible, exploratory and fast. “Most of the time you’re going to your logs, you’re trying to troubleshoot a problem; and today’s problems are different than yesterday’s problems,” Hart stated. “The key to brute force is you’ve got to bring a lot of force as quickly as you possibly can. Our median performance is still better than 100 milliseconds.