I joined Google in 2005. While at Google I've worked in the following areas:Web Search
- Reliability and monitoring of the production web crawling system.
- Improving freshness and comprehensiveness of the Google web search index.
- Building a market economy for provisioning machine resources
- Measuring the statistical behavior of failures at different levels of Google's distributed storage infrastructure to influence storage system and datacenter design.
- Building infrastructure for statistical computing on big data.
- Implementing large-scale traffic forecasting and capacity planning models for web search and related services.
- Summer of Code (In 2005, 2006, 2007, and 2008).
- Contributing to the R Project for Statistical Computing and several open-source R packages.
To help continue bringing in new talent, I have also been fairly involved in the engineering hiring process.
I received an M.Sc. in Mathematics and the Foundations of Computer Science from Oxford University, working with Joel Ouaknine in the area of Software Model Checking. Prior to joining Google I was deeply involved in the FreeBSD operating system, and was the primary release engineer for the FreeBSD 4.x releases.
Poster Paper: Automatic Reconfiguration of Distributed Storage
The 12th International Conference on Autonomic Computing, IEEE (2015), pp. 133-134
Take me to your leader! Online Optimization of Distributed Storage Configurations
Proceedings of the 41st International Conference on Very Large Data Bases, VLDB Endowment (2015), pp. 1490-1501
Janus: Optimal Flash Provisioning for Cloud Storage Workloads
Proceedings of the USENIX Annual Technical Conference, USENIX, Advanced Computing System Association, 2560 Ninth Street, Suite 215, Berkeley, CA 94710, USA (2013), pp. 91-102
Projecting Disk Usage Based on Historical Trends in a Cloud Environment
ScienceCloud 2012 Proceedings of the 3rd International Workshop on Scientific Cloud Computing, ACM, pp. 63-70
Uncertainty in Aggregate Estimates from Sampled Distributed Traces
2012 Workshop on Managing Systems Automatically and Dynamically, USENIX
Large-Scale Parallel Statistical Forecasting Computations in R
Murray Stokely, Farzan Rohani, Eric Tassone
JSM Proceedings, Section on Physical and Engineering Sciences, American Statistical Association, Alexandria, VA (2011)
Availability in Globally Distributed Storage Systems
Proceedings of the 9th USENIX Symposium on Operating Systems Design and Implementation, USENIX (2010)
Using a Market Economy to Provision Compute Resources Across Planet-wide Clusters
Proceedings for the International Parallel and Distributed Processing Symposium 2009, IEEE, pp. 1-8
Parallel Assignments in Software Model Checking
Murray Stokely, Sagar Chaki, Joel Ouaknine
Electr. Notes Theor. Comput. Sci., vol. 157 (2006), pp. 77-94