https://alamrafiul.com/tags/backpressure/ Recent content in Backpressure on Rafiul Alam https://alamrafiul.com/papermod-cover.png https://alamrafiul.com/papermod-cover.png Hugo -- gohugo.io en Fri, 19 Sep 2025 00:00:00 +0000 https://alamrafiul.com/blogs/producer-consumer-bounded/ Fri, 19 Sep 2025 00:00:00 +0000 https://alamrafiul.com/blogs/producer-consumer-bounded/ <h2 id="from-unbounded-to-bounded">From Unbounded to Bounded</h2> <p>In the <a href="https://alamrafiul.com/blogs/producer-consumer-unbounded/">previous article</a>, we explored the unbounded buffer pattern where the queue could grow infinitely. This works until you run out of memory!</p> <p>The <strong>bounded buffer</strong> adds a crucial constraint: <strong>maximum queue size</strong>. This introduces <strong>backpressure</strong> - when the buffer is full, producers must wait for consumers to catch up.</p> <h2 id="why-bounded-buffers-matter">Why Bounded Buffers Matter</h2> <p>Bounded buffers appear everywhere in production systems:</p> <ul> <li><strong>TCP sliding windows</strong> (flow control)</li> <li><strong>HTTP/2 stream flow control</strong> (prevents overwhelm)</li> <li><strong>Message queue limits</strong> (RabbitMQ, Kafka partition limits)</li> <li><strong>Thread pool queues</strong> (bounded task queues)</li> <li><strong>Rate limiters</strong> (token buckets with finite capacity)</li> <li><strong>Circuit breakers</strong> (limit concurrent requests)</li> </ul> <p><strong>The key benefit</strong>: Bounded buffers provide <strong>natural backpressure</strong> and prevent resource exhaustion.</p> https://alamrafiul.com/posts/go-rate-limiter/ Wed, 24 Jul 2024 00:00:00 +0000 https://alamrafiul.com/posts/go-rate-limiter/ Implement rate limiting patterns in Go using token bucket, sliding window, and fixed window algorithms for controlling request rates and preventing system overload.