Peter Lawrey discusses the differences between microservices and monolith architectures, their relative benefits and disadvantages, patterns and strategies implementing low latency microservices.
Martin Thompson focuses on algorithms which provide very high throughput while keeping latency low and predictable, discussing the concurrency theory and implementing these algorithms in Java 8.
Gil Tene provides an in-depth overview of Latency and Response Time Characterization, including proven methodologies for measuring, reporting, and investigating latencies, including pitfalls to avoid.
Charlie Hunt explains the three performance attributes of throughput, latency and (memory) footprint and how each of these are influenced in terms of JVM garbage collection.
Rick Hudson discusses the motivation, performance, and technical challenges of Go's low latency concurrent GC and why the approach fits Go well.
Christopher Meiklejohn looks at applying two techniques together, deterministic data flow programming and conflict-free replicated data types, to create highly available and fault-tolerant systems.
John Davies walks through and demonstrates how to reduce latency while increasing throughput in applications, with demos using Java 8 and lambdas.
Peter Lawrey discusses data-driven reactive systems, profiling latency distribution in such an environment, finding rare bugs, implementing resilience and monitoring.
Brian Troutwine shares insight on using Erlang for a highly concurrent and very low latency bidding system implemented by Adroll.
In this solutions track talk, sponsored by Azul Systems, Gil Tene discusses pitfalls encountered in measuring and characterizing latency, and ways to address them using some new open source tools.
John O’Hara discusses banking business and technology integration, covering: low-latency, high-frequency trading, in-memory caches, multi-terabyte time-series databases, and contracts in NoSQL stores.
Gil Tene introduces and demonstrates how simple and recently open sourced tools can be used to improve and gain higher confidence in both latency measurement and reporting.