Building for Performance with Spring Integration & Spring Batch
summary
Covers using Spring Integration and Spring Batch to architect high-volume, low-latency messaging and integration systems for financial/enterprise environments. Shows architectural patterns and real-world examples of replacing large codebases with Spring Integration configs, integrating large-file processing via Spring Batch, and techniques to maximize throughput while keeping monitoring, error handling and robustness.
AI-written summary · about
chapters (18)
- 0:00Intro
- 1:37Agenda • What is good performance
- 2:33Spring Integration • Implementation of the Enterprise Integration Patterns
- 4:05Zero-impact Code Remoting
- 8:40AMQP Outbound Gateway
- 12:00Client Config
- 12:58The Value of a Millisecond - Financial Services
- 18:41Common Concerns • I'm not in control of its performance -Overhead is low for most scenarios
- 20:41Cost of Admission • Send message to a Queueing Channel • Receive it from the Channel
- 24:49A Failed Metaphor
- 25:49Logic Invocation Cost • What does it cost us to expose existing logic as a transformer?
- 31:43Transactions Are • Typically throughput drops by 2 orders of magnitude
- 34:52JMX • Every component proxied - adds -3.5us to our previous test
- 38:17Per-Message Stats . We can track individual messages by stamping them at key points in the flow
- 43:37Messages Like Monogamy, Threads Do Not . Life is much simpler (& faster) if each message is processed by a single thread
- 45:06Windows Task Manager
- 46:18SI Gateway - Synchronous default-reply-timeout
- 48:56Retail System
related talks
49mBatch Processing in Action
44mBatching for the Modern Enterprise
01h 00mScaling Batch Applications
46mSpring Backend Performance: Techniques Every Dev Should Know
01h 07mJBatch or Not-So-Big Data
01h 11mSpring Integration - Java Configuration and More
44mKeep Spring at Arm's Length!
01h 24mSpring Integration Java DSL