Prevent, Rather than Fix: Scaling to Enable Customers - Jason Ding Keynote

From Velocity Santa Clara 2015: At any company, scale needs to be a priority in order to serve customers of any size. You will have two choices to support the customer’s go-live: (1) Stage the go-live process and fix scale issues while the customer is going through the production phases; or (2) Identify and address scale issues prior to production by simulating peak traffic patterns. It is important to test customer solutions using high fidelity simulations prior to production deployment. The keys to a good simulation are:
1. Representative data and configuration
2. Use cases / business operations
3. Simulation run rules – cache conditions, background noise, simulation time, etc.,
4. Factor for growth based on business projections
5. Iterative enhancement learned from production traffic

This keynote is sponsored by Salesforce.

Jason Ding is senior director of Performance Engineering @ Salesforce. He is currently leading PerfEng teams to work on Salesforce’s next-gen CRM products based on the Lightning platform. He is also leading a cross-functional team to scale the overall Salesforce cloud stack for global enterprise customers. Jason has worked in his 20-year professional career on performance and scalability of microprocessors, server/network systems, OS kernels, on-premise and cloud based enterprise applications. He held technical and management leadership positions from Intel, Cisco, Ariba/SAP before joining Salesforce in 2011. Jason graduated from Texas A&M University with a Ph.D. degree in computer science.