Trouble with Memory / Low latency in applications by Kirk Pepperdine and Joe Sherwin

Nous vous proposons ce mois-ci une soirée performance. Vous y trouverez Joe Sherwin d’Hazelcast qui nous parlera latence, IA et temps réel. Il est suivi de Kirk Pepperdine, alias Monsieur Performance.

This month, we offer you a performance night. You will find Joe Sherwin from Hazelcast who will talk about latency, AI and real-time. He is following by Kirk Pepperdine, aka Mister Performance.

1. Explore the business impact of latency in applications (in English / en Anglais)

Low latency is a gateway for the next generation of computing (aka data processing) and creating applications that were previously unimaginable. Applications which incorporated AI constructs and operate at the edge of the network, all in real time.

In this talk, we will discuss:

  • Application architecture for reducing Application Latency.
  • The business impact of latency in applications.
  • The Hazelcast open-source technology platform for building latency killing applications and processing data streams in real-time & In-memory partitioned data storage.

Présenté par / Presented by: Joe Sherwin

Joe Sherwin is a senior level Solution Architect at Hazelcast® with 18 years of experience in the design, development and implementation of application systems within multi-tier distributed computing environments. Working with clients such as Vanguard, Fannie Mae, Federal Reserve Bank, Citi Group, Bear Stearns, Fixed Income Clearing Corporation, Comcast Corp, Webster Bank, Gartner Group, The Hartford Life Company, IBM Global Services, Mass Mutual, Lincoln National Financial Corporation, Bank of America, and Barnes & Noble Online Group, Mr. Sherwin has been instrumental in the development of large-scale mission critical E-commerce, insurance, and financial systems. He has experience architecting & implementing solution using CORBA, RMI, Java EE compliant distributed Object architectures, in-memory high transaction/low latency solutions using Hazelcast IMDG®, GemFire, Ehcache & Oracle Coherence, and solutions deployable on IaaS or PaaS platforms like Cloud Foundry, Amazon Web Services, Rackspace or Heroku.

2. The Trouble with Memory (in English / en Anglais)

When developers and operations are asked; what is the biggest performance bottleneck you face on a regular basis, it’s rare that memory inefficiency comes up as an answer. Yet our observations suggest that 60% of all application do suffer from this problem. Further more, we see this problem in all versions of Java, Scala, Ruby, Kotlin, Clojure, Groovy, etc. We’ve also seen it in C# running on the CLR, on Android devices, in the cloud, on bare metal systems, just about every where we’ve looked. There are many reasons why this bottleneck is being overlooked. Quite often this memory inefficiency hides itself as lower application throughput and longer more frequent tail latency events meaning very often, other things take the blame. In this session we’ll look at the tell tale signs that your JVM based application is in that 60% and demo steps you can take with your Java application to attack this problem.

Présenté par / Presented by: Kirk Pepperdine

Kirk Pepperdine is Java Champion since 2005 and renowned performance expert. He co-founded JClarity, a startup focused on developing the next generation performance diagnostic tooling. Additionally, Kirk continues to deliver the popular Java Performance Tuning Workshop as well as Java performance tuning services. In addition to working in Java, he has developed and tuned applications written in C/C++, Smalltalk, and a variety of other languages. Kirk has written many articles and spoken often at conferences and user groups on the subject of performance tuning.