FoCS Corporate Connection - "Analyzing Unstructured (BIG) Data to Manage Large IT Environments," Jon Herlocker/CTO for Log Insight and Log Analytics, VMware

Contact Name: 
Jenna Whitney
GDC 6.302
Oct 31, 2013 12:00pm - 1:15pm

Audience: All CS grad students, PhD candidates especially welcome

Abstract: VMware provides software solutions to the worlds largest Enterprise IT environments. Enterprise IT teams are responsible for managing huge numbers of machines, both physical and virtual, sometimes into the hundreds of thousands. These range from massive critical business systems such as email servers or web servers to employee desktops. The diversity of hardware and software platforms that a single organization must support is often mind-boggling. Every one of these individually represents a complex system that must be managed, and when the systems interact with each other, the complexity grows even further. There are endless ways that these complex IT ecosystems can fail.  Failures often happen in cascades, where one failure causes another and another, such that the observed symptoms can be significantly different than the original cause of the cascading failure - what we call the "root cause". Enterprise IT environments require sophisticated software that can help them preemptively identify problems before they cause cascading failures and help them to more rapidly identity and resolve issues when they do occur. In this talk, we will present the challenge, and describe VMware's approach to delivering the solution which consists of a new scale-out data service that enables SQL-database-like queries across unstructured log data and the application of machine learning techniques to help IT admins more quickly identify the trends and root cause of issues.

Bio: Jon Herlocker is the Chief Technology Officer for Log Insight and Log Analytics at VMware. He owns all research & development for the VMware vCenter Log Insight product as well as long term product and technology strategy for log analytics-related products and technologies. In prior roles he was CTO and VP of Engineering of Mozy, a large cloud backup service, and CTO of the EMC Cloud Services Division. He was co-founder of two high-tech startups - MusicStrands (now Strands), and Smart Desktop. He holds a PhD in computer science from the University of Minnesota and was tenured Associate Professor of Computer Science at Oregon State University (Corvallis) where he published a number of highly cited scientific papers and led DARPA-funded research. He was heavily involved in the pioneering work in the area of recommendation systems, and in 2010 he was awarded the Association for Computing Machinery's Software System Award in recognition of developing a software system that has had a lasting influence in the field of computer science. Jon’s research & product experience has spanned distributed systems, machine learning, and human computer interaction.